00578nas a2200193 4500000000100000000000100001008004100002260003100043100001600074700002300090700002100113700002300134700001900157700002500176700001700201700001700218245013500235300001400370 2012 d bSpringer Berlin Heidelberg1 aM. Blachnik1 aPrzemysław Głomb1 aLeszek Rutkowski1 aMarcin Korytkowski1 aRafał Scherer1 aRyszard Tadeusiewicz1 aZadeh Lotfia1 aJacek Zurada00aDo we need complex models for gestures? A comparison of data representation and preprocessing methods for hand gesture recognition a477–48500355nas a2200109 4500000000100000000000100001008004100002100002000043700002300063245014500086022001400231 2012 d1 aMichał Cholewa1 aPrzemysław Głomb00aEstimation of the number of states for gesture recognition with Hidden Markov Models based on the number of critical points in time sequence a0167-865500313nas a2200121 4500000000100000000000100001008004100002100002000043700002300063245008400086300001400170490000700184 2011 d1 aMichał Cholewa1 aPrzemysław Głomb00aGesture data modeling and classification based on critical points approximation a365–3730 v9500607nas a2200241 4500000000100000000000100001008004100002260002600043100002000069700002300089700002100112700002400133700002100157700001900178700001800197700001400215700001500229700001800244700002400262700001800286245003900304020002200343 2013 d bOficyna Wydawnicza PW1 aMichał Cholewa1 aPrzemysław Głomb1 aSebastian Opozda1 aMichał Romaszewski1 aArkadiusz Sochan1 aM. Kosiedowski1 aE. Kuśmierek1 aA. Bęben1 aP. Krawiec1 aA. Stroiński1 aWojciech Burakowski1 aPiotr Krawiec00aAplikacje Sieci Świadomej Treści a978-83-7814-099-300479nas a2200157 4500000000100000000000100001008004100002260002000043100002100063700002300084700002500107700002000132245006300152490000700215520009900222 2009 d bSpringer Verlag1 aJoanna Domańska1 aPrzemysław Głomb1 aPrzemysław Kowalski1 aSławomir Nowak00aModeling of Internet 3D Traffic Using Hidden Markov Models0 v643 a

Artykuł podejmuje temat modelowania złożoności przesyłu siatek progresywnych 3D.

00938nas a2200205 4500000000100000000000100001008004100002260003400043100001700077700002300094700002200117700002200139700002400161700002500185700002200210245005500232300001200287520041100299020002200710 2011 d bSpringeraBerlin / Heidelberg1 aPiotr Gawron1 aPrzemysław Głomb1 aJaroslaw Miszczak1 aZbigniew Puchała1 aTadeusz Czachórski1 aStanisław Kozielski1 aUrszula Stańczyk00aEigengestures for natural human computer interface a49–563 aWe present the application of Principal Component Analysis for data acquired during the design of a natural gesture interface. We investigate the concept of an eigengesture for motion capture hand gesture data and present the visualisation of principal components obtained in the course of conducted experiments. We also show the influence of dimensionality reduction on reconstructed gesture data quality. a978-3-642-23168-100422nas a2200145 4500000000100000000000100001008004100002260004500043100002200088700002400110700002300134700002100157700002000178245007800198 2004 d bWydawnictwa Naukowo-TechniczneaWarszawa1 aKrzysztof Grochla1 aTadeusz Czachórski1 aPrzemysław Głomb1 aArkadiusz Sochan1 aAndrzej Grzywak00aInternet w społeczeństwie informacyjnym - Podstawowe problemy Internetu00285nas a2200121 4500000000100000000000100001000000100002008004100003100002200044700002300066245006700089490000700156 2002 d1 aKrzysztof Grochla1 aPrzemysław Głomb00aKształtowanie ruchu w routerach pracujących w systemie Linux0 v2300741nas a2200145 4500000000100000000000100001008004100002260004500043100002200088700002300110700002100133700002000154245006600174520035500240 2004 d bWydawnictwa Naukowo-TechniczneaWarszawa1 aKrzysztof Grochla1 aPrzemysław Głomb1 aArkadiusz Sochan1 aAndrzej Grzywak00aWspółczesne problemy sieci komputerowych - Nowe technologie3 aRozdział przedstawia budowę i wyniki badań symulacyjnych mechanizmu sterowania transmisją sieciową skalowalnego strumienia wideo. Skalowalność typu SNR otrzymywana jest z wykorzystaniem transformaty falkowej, a sterowanie zrealizowane jest z użyciem regulatora proporcjonalno-różniczkującego w oparciu o pomiar opóźnień kolejnych ramek.00307nas a2200133 4500000000100000000000100001008004100002260000600043100002200049700002100071700002300092700001500115245004300130 2008 d c41 aKrzysztof Grochla1 aArkadiusz Sochan1 aPrzemysław Głomb1 a- Borcochi00aTransmission of scalable video streams00747nas a2200145 4500000000100000000000100001008004100002260004500043100002200088700002100110700002300131700002000154245008000174520034700254 2004 d bWydawnictwa Naukowo-TechniczneaWarszawa1 aKrzysztof Grochla1 aArkadiusz Sochan1 aPrzemysław Głomb1 aAndrzej Grzywak00aWspółczesne problemy sieci komputerowych - Zastosowanie i bezpieczeństwo3 aRozdział prezentuje jednolity system autoryzacji użytkowników umożliwiający dostęp do wszytskich zasobów sieci za pomocą jednego hasła. Zaprezentowano rozwiązanie wdrożone w IITiS PAN. Autorzy wykorzystali oprogramowanie firmy Microsoft oraz pakiety dostępne na zasadach open source takie jak: Samba, Dovecot, Apache oraz smbpamd.00245nas a2200109 4500000000100000000000100001008004100002260001300043100002300056700002100079245003500100 2007 d bSpringer1 aPrzemysław Głomb1 aMarek Kurzyński00aComputer Recognition Systems 200375nas a2200145 4500000000100000000000100001008004100002260003500043100002300078700001800101700002100119700002500140700001700165245004700182 2006 d bAcademic Publishing House EXIT1 aPrzemysław Głomb1 aAndrzej Cader1 aLeszek Rutkowski1 aRyszard Tadeusiewicz1 aJacek Zurada00aArtificial Intelligence and Soft Computing00227nas a2200085 4500000000100000000000100001008004100002100002300043245007500066 2009 d1 aPrzemysław Głomb00aDetection of interest points on 3D data: extending the Harris operator00210nas a2200085 4500000000100000000000100001008004100002100002300043245005800066 2007 d1 aPrzemysław Głomb00aImage Language Terminal Symbols from Feature Analysis00247nas a2200085 4500000000100000000000100001008004100002100002300043245009500066 2005 d1 aPrzemysław Głomb00aAnalysis of fGn and http requests traces using localized multiscale H parameter estimation00310nas a2200109 4500000000100000000000100001008004100002260005100043100002300094700002200117245006100139 2006 d bInstytut Informatyki Teoretycznej i Stosowanej1 aPrzemysław Głomb1 aKrzysztof Grochla00aRepresenting single-link traffic quality with kernel PCA00232nas a2200097 4500000000100000000000100001008004100002100002300043700002200066245004600088 2005 d1 aPrzemysław Głomb1 aKrzysztof Grochla00aModeling Connection State with Kernel PCA00295nas a2200121 4500000000100000000000100001000000100002008004100003100002300044700002200067245007700089490000700166 2002 d1 aPrzemysław Głomb1 aKrzysztof Grochla00aModelowanie stacji roboczych w sieci ATM przy użyciu symulatora OMNeT++0 v2300893nas a2200145 4500000000100000000000100001008004100002260004500043100002300088700002200111700002100133700002000154245007800174520049500252 2004 d bWydawnictwa Naukowo-TechniczneaWarszawa1 aPrzemysław Głomb1 aKrzysztof Grochla1 aArkadiusz Sochan1 aAndrzej Grzywak00aInternet w społeczeństwie informacyjnym - Podstawowe problemy Internetu3 aArtykuł prezentuje przegląd i badania zagadnienia skalowalności w stosowanych współcześnie technologiach kodowania sygnału wideo. W pierwszej części prezentowane są modele skalowalności przyjęte w stosowanych i opracowywanych standardach kompresji, w drugiej przedstawione są wyniki badań jakości transmisji skalowalnych strumieni w różnorodnych warunkach obciążenia. Całość zamknięta jest podsumowaniem możliwości sterowania transmisją sieciową takich strumieni.00287nas a2200097 4500000000100000000000100001008004100002100002300043700002000066245010300086 2008 d1 aPrzemysław Głomb1 aSławomir Nowak00aExperimental rate distortion analysis of image coder with contourlet transform and SPIHT algorithm01225nas a2200169 4500000000100000000000100001000000100002008004100003100002300044700002200067700002100089245008600110300001400196490000700210520081700217022002101034 2007 d1 aPrzemysław Głomb1 aZbigniew Puchała1 aArkadiusz Sochan00aContext selection for efficient bit modeling of contourlet transform coefficients a135–1460 v193 aContourlet transform is named as one possible successor to the wavelet transform as a tool for representation of image information. It has better properties in representation of edge information, which make it suitable for designing image compression systems. It is not clear, however, how to exploit the statistical structure of the coefficients generated by this transform in effective compression algorithm. In this work, authors propose a general, nonparametric method to achieve context map, thus allowing for efficient application of EBCOT/JPEG 2000 bit modeling and encoding approach to contourlet transformed images. The article presents the derivation of the method and experiments performed on a publicly available image set. Keywords: contourlet transform, coefficient bit modeling, context estimation aISSN 1896–533400435nas a2200145 4500000000100000000000100001008004100002260003100043100002300074700002400097700002100121700002100142245008100163250004500244 2012 d bSpringer Berlin Heidelberg1 aPrzemysław Głomb1 aMichał Romaszewski1 aSebastian Opozda1 aArkadiusz Sochan00aChoosing and Modeling the Hand Gesture Database for a Natural User Interface aLecture Notes in Artificial Intelligence00329nas a2200121 4500000000100000000000100001008004100002100002300043700002400066700002100090700002100111245007500132 2011 d1 aPrzemysław Głomb1 aMichał Romaszewski1 aSebastian Opozda1 aArkadiusz Sochan00aChoosing and modeling hand gesture database for natural user interface00386nas a2200133 4500000000100000000000100001008004100002100002300043700002400066700002100090700002100111245011100132490000900243 2011 d1 aPrzemysław Głomb1 aMichał Romaszewski1 aArkadiusz Sochan1 aSebastian Opozda00aUnsupervised Parameter Selection for Gesture Recognition with Vector Quantization and Hidden Markov Models0 v694900297nas a2200121 4500000000100000000000100001008004100002100002300043700002100066245006200087300001200149022001400161 2009 d1 aPrzemysław Głomb1 aKrzysztof Skabek00aLaboratory for the exploration of virtual 3D environments a57–59 a1640-375400520nas a2200193 4500000000100000000000100001008004100002100002300043700002100066700002100087700002300108700001900131700002500150700001600175700001700191245008200208300001400290020002200304 2014 d1 aPrzemysław Głomb1 aArkadiusz Sochan1 aLeszek Rutkowski1 aMarcin Korytkowski1 aRafał Scherer1 aRyszard Tadeusiewicz1 aLotfi Zadeh1 aJacek Zurada00aSurface Mixture Models for the Optimization of Object Boundary Representation a703–714 a978-3-319-07172-501429nas a2200241 4500000000100000000000100001000000100002008004100003100001900044700001800063700001800081700002000099700002300119700002100142700002400163700002100187700001400208700001500222245003900237300001600276520088100292022001401173 2012 d1 aM. Kosiedowski1 aE. Kuśmierek1 aA. Stroiński1 aMichał Cholewa1 aPrzemysław Głomb1 aSebastian Opozda1 aMichał Romaszewski1 aArkadiusz Sochan1 aA. Bęben1 aP. Krawiec00aAplikacje sieci świadomej treści a1491–15023 aArtykuł przedstawia aplikacje korzystające z sieci świadomej treści PI CAN opracowanej w ramach projektu Inżynieria Internetu Przyszłości (IIP)1. Są to trzy aplikacje oferujące dostęp do treści multimedialnych: HomeNetMedia dla celów rozrywki, Rozproszone Wirtualne Muzeum dla edukacji oraz Biblioteka Cyfrowa Pacjenta dla przekazywania informacji medycznej. Scharakteryzowana została funkcjonalność sieci PI CAN z punktu widzenia aplikacji wraz z zasadami współpracy z użytkownikami, tj. dostawcami i konsumentami treści. Poszczególne aplikacje omówione zostały ze wskazaniem sposobu wykorzystania sieci świadomej treści oraz implementacji usług zewnętrznych powiązanych z siecią PI CAN specyficznych dla aplikacji. Przedstawione zostały również wyniki integracji poszczególnych aplikacji z siecią PI CAN oraz wyniki przeprowadzonych testów. a1230-349601032nas a2200193 4500000000100000000000100001000000100002008004100003260000700044100002000051700002100071700001800092700002300110245008500133300001400218490000700232520058500239022001400824 2010 d c111 aSławomir Nowak1 aJoanna Domańska1 aMateusz Nowak1 aPrzemysław Głomb00aProgressive 3D meshes transmission: traffic generating and simulation evaluation a247–2590 v223 aThe paper presents issues related to the progressive transmission of 3D meshes in clientserver architecture. A tool generating network traffic for different 3D scenes and scenarios is presented. The objective was to evaluate the transmission dynamics and possible impact of 3D traffic on the amount of data available in VR browsing client, which relates to browsing QoE (Quality of Experience). Simulation studies were carried out and HMM model of traffic generated by 3D server was worked out, planned for use in modeling works on future internet protocols, servicing 3D traffic. a1896-533400389nas a2200109 4500000000100000000000100001008004100002100002000043700002300063245010600086520008700192 2010 d1 aSławomir Nowak1 aPrzemysław Głomb00aREMOTE VIRTUAL REALITY: EXPERIMENTAL INVESTIGATIONSOF PROGRESSIVE 3D MESH TRANSMISSION IN IP NETWORKS3 aPerformance Modelling and Evaluation of Heterogeneous Networks, proc. HETNET 2010.00337nas a2200121 4500000000100000000000100001008004100002100002400043700001700067700002300084245007200107856003600179 2011 d1 aMichał Romaszewski1 aPiotr Gawron1 aPrzemysław Głomb00aTowards a natural gesture interface: LDA-based gesture separability uhttps://arxiv.org/abs/1109.503400362nas a2200133 4500000000100000000000100001008004100002260001900043100002400062700002300086245009900109300001400208490000600222 2011 d c2011bSpringer1 aMichał Romaszewski1 aPrzemysław Głomb00aThe Effect of Multiple Training Sequences on HMM Classification of Motion Capture Gesture Data a365–3730 v400332nas a2200145 4500000000100000000000100001008004100002260000600043100002400049700002300073245005200096300001200148490001200160022001400172 2009 d cV1 aMichał Romaszewski1 aPrzemysław Głomb00a3D Mesh Approximation Using Vector Quantization a71–780 v57/2009 a1615-387100380nas a2200145 4500000000100000000000100001008004100002260001300043100002400056700002300080700001700103700001900120245008100139300001400220 2014 d c10bIEEE1 aMichał Romaszewski1 aPrzemysław Głomb1 aPiotr Gawron1 aKhalifa Djemal00aNatural hand gestures for human identification in a Human-Computer Interface a404–40900432nas a2200157 4500000000100000000000100001008004100002260003400043100002400077700002100101700002300122700002000145700002400165245007100189300001400260 2009 d bAcademy Publishing House EXIT1 aMichał Romaszewski1 aSebastian Opozda1 aPrzemysław Głomb1 aRyszard Choraś1 aAntoni Zabłudowski00a3D Mesh Identification Using Random Walks and Hidden Markov Models a302–30800771nas a2200145 4500000000100000000000100001008004100002260004500043100002100088700002200109700002300131700002000154245007800174520037300252 2004 d bWydawnictwa Naukowo-TechniczneaWarszawa1 aArkadiusz Sochan1 aKrzysztof Grochla1 aPrzemysław Głomb1 aAndrzej Grzywak00aInternet w społeczeństwie informacyjnym - Podstawowe problemy Internetu3 aW rozdziale zarysowano koncepcję usług katalogowych i metody dostępu do nich. Przedstawiono architekturę aplikacji WWW i różne rozwiązania stosowane do tworzenia takich aplikacji. Zaprezentowano wyniki badań porównawczych przeprowadzonych dla tych rozwiązań. Na koniec scharakteryzowano opracowaną aplikację WWW korzystającą z usług katalogowych i XML.00774nas a2200145 4500000000100000000000100001008004100002260004500043100002100088700002300109700002200132700002000154245006600174520038800240 2004 d bWydawnictwa Naukowo-TechniczneaWarszawa1 aArkadiusz Sochan1 aPrzemysław Głomb1 aKrzysztof Grochla1 aAndrzej Grzywak00aWspółczesne problemy sieci komputerowych - Nowe technologie3 aW rozdziale opisano prace badawcze dotyczące transmisji strumieni MPEG-2 poprzez kanał o zmiennej przepływności bitowej w sieci ATM. Prace miały charakter zarówno doświadczalno-pomiarowy, jak i analityczny. Określono zależność funkcyjną pomiędzy parametrami ruchu sieci ATM, natężeniem strumienia wideo a jakością transmisji multimedialnej. Zależność powiązano z00499nas a2200181 4500000000100000000000100001008004100002100002100043700002300064700002100087700002400108700002100132700002200153700001400175700001300189245010100202300001400303 2011 d1 aArkadiusz Sochan1 aPrzemysław Głomb1 aKrzysztof Skabek1 aMichał Romaszewski1 aSebastian Opozda1 aAndrzej Kwiecień1 aPiotr Gaj1 aP. Stera00aVirtual Museum as an Example of 3D Content Distribution in the Architecture of a Future Internet a459–46400402nas a2200169 4500000000100000000000100001008004100002100001700043700001800060700001200078700001600090700001900106700001500125700002300140245005500163300001400218 2009 d1 aE. Switonski1 aD. Tejszerska1 aM. Gzik1 aW. Wolanski1 aB. Gzik-Zroska1 aM. Mandera1 aPrzemysław Głomb00aInteraktywna kinezyterapia dzieci z płaskostopiem a211–21400280nas a2200097 4500000000100000000000100001008004100002100002300043700002000066245009600086 2009 d1 aPrzemysław Głomb1 aSławomir Nowak00aIMAGE CODING WITH CONTOURLET / WAVELET TRANSFORMS AND SPIHT ALGORITHM:AN EXPERIMENTAL STUDY01043nas a2200169 4500000000100000000000100001008004100002260004300043100001700086700002300103700002100126700002200147700002000169245006900189520059700258020001800855 2005 d bWydawnictwa Komunikacji i Łączności1 aT. Białecki1 aPrzemysław Głomb1 aArkadiusz Sochan1 aAndrzej Kwiecień1 aAndrzej Grzywak00aWysokowydajne sieci komputerowe - Zastosowanie i bezpieczeństwo3 aW rozdziale przedstawiono możliwości wykorzystania standardu JPEG-2000 do kodowania strumieni przesyłanych przez kanał transmisyjny o wysokim współczynniku błędu i start. Scharakteryzowano metodę kodowania i wynikające z niej cechy skalowalności. W dalszej części opisano zastosowany model i przeprowadzone badania dotyczące stabilizacji jakości prezentowanych sekwencji obrazów z wykorzystaniem informacji nadmiarowej. Pokazano, że kodowanie obrazów ruchomych oparte na warstwach i płaszczyznach bitowych w połączeniu z kodami nadmiarowymi poprawia jakość prezentacji. a83-206-1573-900923nas a2200169 4500000000100000000000100001008004100002260008200043100001700125700002300142700002100165700002200186700002000208245006000228520044700288020001800735 2005 d bWydawnictwa Komunikacji i ŁącznościaWydawnictwa Komunikacji l Łaczności1 aT. Białecki1 aPrzemysław Głomb1 aArkadiusz Sochan1 aAndrzej Kwiecień1 aKazimiera Wódz00aTechniczne i społeczne problemy zastosowania Internetu3 aW rozdziale przedstawiono analizę mechanizmu utrzymania jakości prezentacji strumienia obrazów JPEG 2000 w warunkach transmisji z utratą danych z wykorzystaniem dodatkowej informacji nadmiarowej. Kolejne podrozdziały przedstawiają: teoretyczną analizę zagadnienia skalowanej transmisji wideo w środowisku o wysokiej stopie błędu z wykorzystaniem informacji korekcyjnej, opracowane stanowisko badawcze, wyniki badań i podsumowanie. a83-206-1590-901428nas a2200181 4500000000100000000000100001008004100002260000900043100002100052700002300073700002700096245009000123856010300213300001200316490000700328520089500335022001601230 2015 d c20151 aKrzysztof Domino1 aPrzemysław Głomb1 aZbigniew Łaskarzewski00aClassification of LPG clients using the Hurst exponent and the correlation coeficient uhttp://yadda.icm.edu.pl/yadda/element/bwmeta1.element.baztech-1ba91297-f921-4524-927c-381996c36512 a13–240 v273 a

In this paper we present the analysis of the gas usage for different types of buildings. First, we introduce the classical theory of building heating. This allows the establishment of theoretical relations between gas consumption time series and the outside air temperature for different types of buildings, residential and industrial. These relations imply dierent auto-correlations of gas usage time series as well as different cross-correlations between gas consumption and temperature time series for different types of buildings. Therefore, the autocorrelation and the cross-correlation were used to classify the buildings into three classes: housing, housing with high thermal capacity, and industry. The Hurst exponent was calculated using the global DFA to investigate auto-correlation, while the Kendall's τ rank coeficient was calculated to investigate cross-correlation

 a 1896-5334 00592nas a2200217 4500000000100000000000100001008004100002260003800043100002300081700002000104700002100124700002300145700001800168700002500186700001600211700001700227245008700244300001200331490000900343020002200352 2015 d bSpringer International Publishing1 aPrzemysław Głomb1 aMichał Cholewa1 aLeszek Rutkowski1 aMarcin Korytkowski1 aRafal Scherer1 aRyszard Tadeusiewicz1 aLotfi Zadeh1 aJacek Zurada00aExperimental Evaluation of Selected Approaches to Covariance Matrix Regularization a391-4010 v9120 a978-3-319-19368-700315nas a2200109 4500000000100000000000100001008004100002100002400043700002300067245009300090020002200183 2016 d1 aMichał Romaszewski1 aPrzemysław Głomb00aParameter Estimation for HOSVD-based Approximation of Temporally Coherent Mesh Sequences a978-989-758-175-500491nas a2200157 4500000000100000000000100001008004100002100002400043700002300067700002000090245011800110856007100228300001200299490000800311022001400319 2016 d1 aMichał Romaszewski1 aPrzemysław Głomb1 aMichał Cholewa00aSemi-supervised hyperspectral classification from a small number of training samples using a co-training approach uhttp://www.sciencedirect.com/science/article/pii/S0924271616303446 a60 - 760 v121 a0924-271600274nas a2200109 4500000000100000000000100001008004100002100002000043700002300063245005600086020002200142 2016 d1 aMichał Cholewa1 aPrzemysław Głomb00aTwo Stage SVM Classification for Hyperspectral Data a978-989-758-173-100288nas a2200109 4500000000100000000000100001008004100002100002300043700002000066245007000086020002200156 2016 d1 aPrzemysław Głomb1 aMichał Cholewa00aPerformance of Interest Point Descriptors on Hyperspectral Images a978-989-758-175-501088nas a2200181 4500000000100000000000100001008004100002100002000043700001700063700002300080700001900103245007200122856004800194300000800242490000700250520063500257022001400892 2017 d1 aMichał Cholewa1 aPiotr Gawron1 aPrzemysław Głomb1 aDariusz Kurzyk00aQuantum hidden Markov models based on transition operation matrices uhttp://dx.doi.org/10.1007/s11128-017-1544-8 a1010 v163 a

In this work, we extend the idea of quantum Markov chains (Gudder in J Math Phys 49(7):072105 [3]) in order to propose quantum hidden Markov models (QHMMs). For that, we use the notions of transition operation matrices and vector states, which are an extension of classical stochastic matrices and probability distributions. Our main result is the Mealy QHMM formulation and proofs of algorithms needed for application of this model: Forward for general case and Vitterbi for a restricted class of QHMMs. We show the relations of the proposed model to other quantum HMM propositions and present an example of application.

 a1573-133200522nas a2200169 4500000000100000000000100001008004100002260001200043100002300055700002400078700002000102700002100122245011600143856007100259490000800330022001400338 2018 d c07/20181 aPrzemysław Głomb1 aMichał Romaszewski1 aMichał Cholewa1 aKrzysztof Domino00aApplication of hyperspectral imaging and Machine Learning methods for the detection of gunshot residue patterns uhttp://www.sciencedirect.com/science/article/pii/S03790738183036570 v290 a0379-073801191nas a2200145 4500000000100000000000100001008004100002260003800043100002400081700002300105700002000128245009700148520077800245020002201023 2018 d bSpringer International Publishing1 aMichał Romaszewski1 aPrzemysław Głomb1 aMichał Cholewa00aAdaptive, Hubness-Aware Nearest Neighbour Classifier with Application to Hyperspectral Data3 a

We present an extension of the Nearest Neighbour classifier that can adapt to sample imbalances in local regions of the dataset. Our approach uses the hubness statistic as a measure of a relation between new samples and the existing training set. This allows to estimate the upper limit of neighbours that vote for the label of the new instance. This estimation improves the classifier performance in situations where some classes are locally under-represented. The main focus of our method is to solve the problem of local undersampling that exists in hyperspectral data classification. Using several well-known Machine Learning and hyperspectral datasets, we show that our approach outperforms standard and distance-weighted kNN, especially for high values of k.

 a978-3-030-00840-600427nas a2200157 4500000000100000000000100001008004100002260001000043100002000053700002300073700002400096245011600120300001200236490000700248022001400255 2019 d cMarch1 aMichał Cholewa1 aPrzemysław Głomb1 aMichał Romaszewski00aA Spatial-Spectral Disagreement-Based Sample Selection With an Application to Hyperspectral Data Classification a467-4710 v16 a1545-598X02068nas a2200181 4500000000100000000000100001008004100002100002200043700002300065700002300088700001700111245006100128856007100189300001100260490000700271520159400278022001401872 2018 d1 aBartosz Grabowski1 aWojciech Masarczyk1 aPrzemysław Głomb1 aAgata Mendys00aAutomatic pigment identification from hyperspectral data uhttp://www.sciencedirect.com/science/article/pii/S1296207417306544 a1 - 120 v313 a

Art objects conservation or historical analysis necessitates a thorough knowledge of materials used by the artist and their subsequent changes. In the case of paintings this requires the ability to correctly identify the pigments that were used for creation or later restoration of the artwork. This is a challenging problem, as the applied method should be non-contact, robust for the wide variety of chemical substances used and straightforward in the interpretation. Recently, the hyperspectral imaging has emerged as a promising measuring methodology for this kind of the artwork analysis; the combination of acquiring spectral information and planar (photography-like) pixel arrangement provides a lot of potential for material characterization. While initial studies of hyperspectral imaging application to art objects analysis are encouraging, the difficulties of working with its multidimensional data are acknowledged; in many cases complex algorithms are required to fully utilize its potential. In this paper, we study the problem of algorithm design for pigment identification based on a hyperspectral image of a painting. We combine various processing steps to achieve a robust solution requiring minimal user intervention. Using a special set of paintings and a reference pigment database we demonstrate the viability of applying this method in the pigment recognition setting. Our results confirm the potential of using hyperspectral imaging in the art conservation setting, and based on them we discuss the potential construction and elements of such an algorithm.

 a1296-207402286nas a2200157 4500000000100000000000100001008004100002100002300043700002300066700002200089700002400111245012800135856003700263490000700300520182100307 2020 d1 aWojciech Masarczyk1 aPrzemysław Głomb1 aBartosz Grabowski1 aMateusz Ostaszewski00aEffective Training of Deep Convolutional Neural Networks for Hyperspectral Image Classification through Artificial Labeling uhttps://arxiv.org/abs/1909.055070 v123 a

Hyperspectral imaging is a rich source of data, allowing for a multitude of effective applications. However, such imaging remains challenging because of large data dimension and, typically, a small pool of available training examples. While deep learning approaches have been shown to be successful in providing effective classification solutions, especially for high dimensional problems, unfortunately they work best with a lot of labelled examples available. The transfer learning approach can be used to alleviate the second requirement for a particular dataset: first the network is pre-trained on some dataset with large amount of training labels available, then the actual dataset is used to fine-tune the network. This strategy is not straightforward to apply with hyperspectral images, as it is often the case that only one particular image of some type or characteristic is available. In this paper, we propose and investigate a simple and effective strategy of transfer learning that uses unsupervised pre-training step without label information. This approach can be applied to many of the hyperspectral classification problems. The performed experiments show that it is very effective at improving the classification accuracy without being restricted to a particular image type or neural network architecture. The experiments were carried out on several deep neural network architectures and various sizes of labeled training sets. The greatest improvement in overall accuracy on the Indian Pines and Pavia University datasets is over 21 and 13 percentage points, respectively. An additional advantage of the proposed approach is the unsupervised nature of the pre-training step, which can be done immediately after image acquisition, without the need of the potentially costly expert’s time.

00555nas a2200169 4500000000100000000000100001008004100002260007300043100002300116700002100139700002400160700002000184245011100204856003700315300001100352020002200363 2019 d c2019bWroclaw University of Science and TechnologyaWrocław, Poland1 aPrzemysław Głomb1 aKrzysztof Domino1 aMichał Romaszewski1 aMichał Cholewa00aBand selection with Higher Order Multivariate Cumulants for small target detection in hyperspectral images uhttps://arxiv.org/abs/1808.03513 ap. 121 a978-83-943803-2-800446nas a2200145 4500000000100000000000100001008004100002260006700043100002200110700002300132700002400155700002400179245007500203020002200278 2019 d bWroclaw University of Science and TechnologyaWrocław, Poland1 aBartosz Grabowski1 aPrzemysław Głomb1 aMichał Romaszewski1 aMateusz Ostaszewski00aUnsupervised deep learning approach to hyperspectral anomaly detection a978-83-943803-2-800409nas a2200133 4500000000100000000000100001008004100002260001300043100002300056700002400079245006100103856008900164020002200253 2020 d bElsevier1 aPrzemysław Głomb1 aMichał Romaszewski00aAnomaly detection in hyperspectral remote sensing images uhttps://www.elsevier.com/books/hyperspectral-remote-sensing/pandey/978-0-08-102894-0 a978-0-08-102894-003407nas a2200193 4500000000100000000000100001008004100002260001200043100002000055700002400075700002300099700002200122700002000144245008300164856004600247490000700293520289900300022001403199 2020 d c11/20201 aKamil Książek1 aMichał Romaszewski1 aPrzemysław Głomb1 aBartosz Grabowski1 aMichał Cholewa00aBlood Stain Classification with Hyperspectral Imaging and Deep Neural Networks uhttps://www.mdpi.com/1424-8220/20/22/66660 v203 a

In recent years, growing interest in deep learning neural networks has raised a question on how they can be used for effective processing of high-dimensional datasets produced by hyperspectral imaging (HSI). HSI, traditionally viewed as being within the scope of remote sensing, is used in non-invasive substance classification. One of the areas of potential application is forensic science, where substance classification on the scenes is important. An example problem from that area—blood stain classification—is a case study for the evaluation of methods that process hyperspectral data. To investigate the deep learning classification performance for this problem we have performed experiments on a dataset which has not been previously tested using this kind of model. This dataset consists of several images with blood and blood-like substances like ketchup, tomato concentrate, artificial blood, etc. To test both the classic approach to hyperspectral classification and a more realistic application-oriented scenario, we have prepared two different sets of experiments. In the first one, Hyperspectral Transductive Classification (HTC), both a training and a test set come from the same image. In the second one, Hyperspectral Inductive Classification (HIC), a test set is derived from a different image, which is more challenging for classifiers but more useful from the point of view of forensic investigators. We conducted the study using several architectures like 1D, 2D and 3D convolutional neural networks (CNN), a recurrent neural network (RNN) and a multilayer perceptron (MLP). The performance of the models was compared with baseline results of Support Vector Machine (SVM). We have also presented a model evaluation method based on t-SNE and confusion matrix analysis that allows us to detect and eliminate some cases of model undertraining. Our results show that in the transductive case, all models, including the MLP and the SVM, have comparative performance, with no clear advantage of deep learning models. The Overall Accuracy range across all models is 98–100% for the easier image set, and 74–94% for the more difficult one. However, in a more challenging inductive case, selected deep learning architectures offer a significant advantage; their best Overall Accuracy is in the range of 57–71%, improving the baseline set by the non-deep models by up to 9 percentage points. We have presented a detailed analysis of results and a discussion, including a summary of conclusions for each tested architecture. An analysis of per-class errors shows that the score for each class is highly model-dependent. Considering this and the fact that the best performing models come from two different architecture families (3D CNN and RNN), our results suggest that tailoring the deep neural network architecture to hyperspectral data is still an open problem.

 a1424-822001424nas a2200145 4500000000100000000000100001008004100002100002400043700002300067700002100090700002000111245008000131856004300211520102400254 2020 d1 aMichał Romaszewski1 aPrzemysław Głomb1 aArkadiusz Sochan1 aMichał Cholewa00aA Dataset for Evaluating Blood Detection in Hyperspectral Images [Data set] uhttps://doi.org/10.5281/zenodo.39849053 a

The sensitivity of hyperspectral imaging (imaging spectroscopy) to haemoglobin derivatives makes it a promising tool for detection and classification of blood. However, due to complexity and high dimensionality of hyperspectral images, the development of hyperspectral blood detection algorithms is challenging. To facilitate their development, we present a new hyperspectral blood detection dataset. This dataset consists of 14 hyperspectral images (ENVI format) of a mock-up scene containing blood and visually similar substances (e.g. artificial blood or tomato concentrate). Images were taken over a period of three weeks and differ in terms of background composition and lighting intensity. To facilitate the use of data, the dataset includes an annotation of classes: pixels where blood and similar substances are visible have been marked by the authors. The main intention behind the dataset is to serve as testing data for Machine Learning methods for hyperspectral target detection and classification.

01436nas a2200169 4500000000100000000000100001008004100002100002400043700002300067700002100090700002000111245006900131856003700200490000800237520100700245022001401252 2021 d1 aMichał Romaszewski1 aPrzemysław Głomb1 aArkadiusz Sochan1 aMichał Cholewa00aA dataset for evaluating blood detection in hyperspectral images uhttps://arxiv.org/abs/2008.102540 v3203 a

The sensitivity of imaging spectroscopy to haemoglobin derivatives makes it a promising tool for detecting blood. However, due to complexity and high dimensionality of hyperspectral images, the development of hyperspectral blood detection algorithms is challenging. To facilitate their development, we present a new hyperspectral blood detection dataset. This dataset, published under an open access license, consists of multiple detection scenarios with varying levels of complexity. It allows to test the performance of Machine Learning methods in relation to different acquisition environments, types of background, age of blood and presence of other blood-like substances. We have explored the dataset with blood detection experiments, for which we have used a hyperspectral target detection algorithm based on the well-known Matched Filter detector. Our results and their discussion highlight the challenges of blood detection in hyperspectral data and form a reference for further works.

 a0379-073801481nas a2200205 4500000000100000000000100001008004100002260004200043100002000085700002300105700002400128700002000152700002200172700001900194245010400213856003700317490001000354520088900364022002201253 2022 d c05/2022bSpringer, ChamaLecce, Italy1 aKamil Książek1 aPrzemysław Głomb1 aMichał Romaszewski1 aMichał Cholewa1 aBartosz Grabowski1 aKrisztian Buza00aImproving Autoencoder Training Performance for Hyperspectral Unmixing with Network Reinitialisation uhttps://arxiv.org/abs/2109.137480 v132313 a

Neural networks, in particular autoencoders, are one of the most promising solutions for unmixing hyperspectral data, i.e. reconstructing the spectra of observed substances (endmembers) and their relative mixing fractions (abundances), which is needed for effective hyperspectral analysis and classification. However, as we show in this paper, the training of autoencoders for unmixing is highly dependent on weights initialisation; some sets of weights lead to degenerate or low-performance solutions, introducing negative bias in the expected performance. In this work, we experimentally investigate autoencoders stability as well as network reinitialisation methods based on coefficients of neurons’ dead activations. We demonstrate that the proposed techniques have a positive effect on autoencoder training in terms of reconstruction, abundances and endmembers errors.

 a978-3-031-06427-201607nas a2200193 4500000000100000000000100001008004100002260001200043100002200055700001700077700001900094700002300113700002000136700002700156245006100183856005900244490000700303520110300310 2022 d c11/20221 aKrzysztof Grochla1 aAnna Strzoda1 aRafał Marjasz1 aPrzemysław Głomb1 aKamil Książek1 aZbigniew Łaskarzewski00aEnergy-Aware Algorithm for Assignment of Relays in LPWAN uhttps://dl.acm.org/doi/10.1145/3544561?cid=996599133970 v183 a

The Low Power Wide Area Networks allow maintaining connectivity with devices over a distance of a few kilometres. However, due
to the presence of obstacles the communication range in an urban environment may be shorter, which creates multiple blind spots or
areas with limited coverage. The nodes acting as relays may solve this problem by forwarding the data from other devices with poor
connectivity. We propose and investigate a novel algorithm that assigns devices to the relay role while keeping the battery constraints.
We define the algorithm, describe the simulation testbed, and present performance evaluation in several scenarios, including real-life
topologies of IoT networks. The proposed relay selection scheme allows increasing the lifetime of the network by selecting the relay
taking into account the current battery capacity. Additionally, it outperforms a previously proposed relay selection algorithm for LP
WAN, showing an average 4% decrease in transmission time and an order of magnitude lower computation complexity.

02153nas a2200181 4500000000100000000000100001008004100002260004900043100002200092700002300114700002000137700001900157245008500176856006600261490000900327520161300336022002201949 2022 d c11/2022bSpringeraHo Chi Minh City, Vietnam1 aBartosz Grabowski1 aPrzemysław Głomb1 aKamil Książek1 aKrisztian Buza00a Improving Autoencoders Performance for Hyperspectral Unmixing Using Clustering uhttps://link.springer.com/chapter/10.1007/978-981-19-8234-7_90 v17163 a

Hyperspectral cameras acquire images containing information across the electromagnetic spectrum, which convey useful information about the scene. To enable effective analysis of such data, spectral unmixing is often used. It is an important task in hyperspectral imaging, allowing one to obtain the information about spectral endmembers which make up each hyperspectral pixel. This task, traditionally solved with dedicated statistical methods, has recently been explored with deep learning methods. One of the methods well-suited to this task are autoencoders. These neural networks are initialized using multiple random weights, and their initialization often has a significant impact on their efficiency. Because of that, to improve the initialization of autoencoders for the spectral unmixing task, we propose to use the pre-training scheme consisting of clustering-based artificial labeling. We test the approach on two popular hyperspectral datasets, i.e. Samson and Jasper Ridge. Our experiment delivers promising results, improving autoencoders effectiveness in the case of Samson dataset, i.e. for 25-class labeling endmembers’ and abundances’ errors improve by 0.045 and 0.008, respectively. The worse results in the case of Jasper Ridge dataset (improvement of the endmembers’ error by 0.001, and worsening of the abundances’ error by 0.006 for 25-classes labeling) show that more research is required to understand when the proposed approach improves the results of the spectral unmixing. The auxiliary experiments that we also conduct allow us to partially answer that question.

 a978-981-19-8233-000340nas a2200133 4500000000100000000000100001008004100002260003800043100001700081700002200098700002300120700001800143245004500161 2023 d c07/2023bIEEEaMarrakesh, Morocco1 aAnna Strzoda1 aKrzysztof Grochla1 aPrzemysław Głomb1 aAndrzej Madej00aLink failure prediction in LoRa networks01389nas a2200193 4500000000100000000000100001008004100002260001200043100002300055700002000078700001900098700001800117700002400135245006600159856004000225490000700265520090900272022001401181 2023 d c05/20231 aPrzemysław Głomb1 aMichał Cholewa1 aWojciech Koral1 aAndrzej Madej1 aMichał Romaszewski00aDetection of emergent leaks using machine learning approaches uhttps://doi.org/10.2166/ws.2023.1180 v233 a

In this work, we focus on the detection of leaks occurring in district metered areas (DMAs). Those leaks are observable as a number of time-related deviations from zone patterns over days or weeks. While they are detectable given enough time, due to the huge cost of water loss resulting from an undetected leak, the main challenge is to find them as soon as possible, when the deviation from the zone pattern is small. Using our collected observational data, we investigate the appearance of leaks and discuss the performance of several machine learning (ML) anomaly detectors in detecting them. We test a diverse set of six anomaly detectors, each based on a different ML algorithm, on nine scenarios containing leaks and anomalies of various kinds. The proposed approach is very effective at quickly (within hours) identifying the presence of a leak, with a limited number of false positives.

 a1606-974902406nas a2200241 4500000000100000000000100001008004100002260001200043100002000055700002200075700002300097700002400120700001800144700001900162700001900181700001700200700001900217700001900236700002100255245017800276856006600454520164400520 2023 d c08/20231 aKamil Książek1 aWilhelm Masarczyk1 aPrzemysław Głomb1 aMichał Romaszewski1 aIga Stokłosa1 aPiotr Ścisło1 aPaweł Dębski1 aRobert Pudlo1 aKrisztian Buza1 aPiotr Gorczyca1 aMagdalena Piegza00aThe analysis of heart rate variability and accelerometer mobility data in the assessment of symptom severity in psychotic disorder patients using a wearable Polar H10 sensor uhttps://www.medrxiv.org/content/10.1101/2023.08.04.23293640v13 a

Background and Objective: Advancement in mental health care requires easily accessible diagnostic and treatment assessment tools. There is an ongoing search for biomarkers that would enable objectification and automatization of the diagnostic and treatment process dependent on a psychiatric interview. Current wearable technology and computational methods make it possible to incorporate heart rate variability (HRV), an indicator of autonomic nervous system functioning and a potential biomarker of disease severity in mental disorders, into accessible diagnostic and treatment assessment frameworks.
Methods: We used a commercially available electrocardiography (ECG) chest strap with a built-in accelerometer, i.e. Polar H10, to record R-R intervals and activity of 30 hospitalized schizophrenia or bipolar disorder patients and 30 control participants for 1.5-2 hours time periods.
We performed an analysis to assess the relationship between HRV and the Positive and Negative Syndrome Scale (PANSS) test scores. The source code for the reproduction of all experiments is available on GitHub while the dataset is available in Zenodo.
Results and Conclusions: Mean HRV values were lower in the patient group and negatively correlated with the results of the PANSS general subcategory. For the control group, we also discovered the inversely proportional dependency between the mobility coefficient based on accelerometer data and HRV. This relationship was less pronounced for the patient group. This indicates that HRV and mobility may be promising markers in disease diagnosis.

01717nas a2200217 4500000000100000000000100001008004100002100002000043700002200063700002300085700002400108700001800132700001900150700001900169700001700188700001900205700002100224245014000245856003800385520107600423 2023 d1 aKamil Książek1 aWilhelm Masarczyk1 aPrzemysław Głomb1 aMichał Romaszewski1 aIga Stokłosa1 aPiotr Ścisło1 aPaweł Dębski1 aRobert Pudlo1 aPiotr Gorczyca1 aMagdalena Piegza00aHRV-ACC: a dataset with R-R intervals and accelerometer data for the diagnosis of psychotic disorders using a Polar H10 wearable sensor uhttps://zenodo.org/record/81712663 a

The issue of diagnosing psychotic diseases, including schizophrenia and bipolar disorder, in particular, the objectification of symptom severity assessment, is still a problem requiring the attention of researchers. Two measures that can be helpful in patient diagnosis are heart rate variability calculated based on electrocardiographic signal and accelerometer mobility data. The following dataset contains data from 30 psychiatric ward patients having schizophrenia or bipolar disorder and 30 healthy persons. The duration of the measurements for individuals was usually between 1.5 and 2 hours. R-R intervals necessary for heart rate variability calculation were collected simultaneously with accelerometer data using a wearable Polar H10 device. The Positive and Negative Syndrome Scale (PANSS) test was performed for each patient participating in the experiment, and its results were attached to the dataset. Furthermore, the code for loading and preprocessing data, as well as for statistical analysis, was included on the corresponding GitHub repository.

00488nas a2200157 4500000000100000000000100001008004100002100002400043700002100067700002400088700002300112700002600135700002000161245013500181022001400316 2023 d1 aAgnieszka Ciemięga1 aKatarzyna Maresz1 aMichał Romaszewski1 aPrzemysław Głomb1 aPaulina Krupska-Wolas1 aKrystian Prusik00aHyperspectral Imaging as a Facile and Non-Destructive Method for Size Analysis of Gold Nanoparticles Deposited on Porous Materials a0934-086601285nas a2200205 4500000000100000000000100001008004100002260015900043100001900202700002000221700002200241700002300263700001900286700002100305245012300326856004500449490000900494520055400503022002201057 2023 d c09/2023bKnižnicné a edicné centrum, Fakulta matematiky, fyziky a informatiky, Univerzita Komenského, Mlynská dolina, BratislavaaTatranské Matliare1 aKrisztian Buza1 aKamil Książek1 aWilhelm Masarczyk1 aPrzemysław Głomb1 aPiotr Gorczyca1 aMagdalena Piegza00aA Simple and Effective Classifier for the Detection of Psychotic Disorders based on Heart Rate Variability Time Series uhttps://ceur-ws.org/Vol-3498/paper28.pdf0 v34983 a

In this paper, we focus on automated detection of schizophrenia and bipolar disorder. For this task, we describe a simple and effective classifier, i.e. convolutional nearest neighbor. It provides a data-driven and objective approach for the detection of schizophrenia and bipolar disorder based on heart rate variability time series. According to our results, our approach is able to distinguish whether the selected person belongs to the patient group with an accuracy of 85% and area under receiver-operator characteristic curve of 0.92.

 a978-80-8147-132-200633nas a2200229 4500000000100000000000100001008004100002100002300043700002300066700002500089700002200114700002300136700001900159700001800178700002400196700001700220700002500237700002200262700002100284700002200305245007600327 2023 d1 aBartłomiej Gardas1 aPrzemysław Głomb1 aPrzemysław Sadowski1 aZbigniew Puchała1 aKonrad Jałowiecki1 aŁukasz Pawela1 aOrphee Faucoz1 aPierre-Marie Brunet1 aPiotr Gawron1 aMatthijs Van Waveren1 aMickaël Savinaud1 aGuillaume Pasero1 aVeronique Defonte00aHyper-Spectral Image Classification Using Adiabatic Quantum Computation00491nas a2200169 4500000000100000000000100001008004100002260004000043100002300083700002400106700002000130700001900150700001800169700002000187700002500207245008900232 2023 d bWydawnictwo Politechniki Łódzkiej1 aPrzemysław Głomb1 aMichał Romaszewski1 aMichał Cholewa1 aWojciech Koral1 aAndrzej Madej1 aMaciej Skrabski1 aKatarzyna Kołodziej00aMachine Learning for Water Leak Detection and Localization in the WaterPrime Project00544nas a2200181 4500000000100000000000100001008004100002100002300043700002000066700001900086700001700105700001700122700002100139700002100160700001900181245012300200856003900323 2023 d1 aPrzemysław Głomb1 aMichał Cholewa1 aPaweł Foszner1 aJakub Bularz1 aMaria Ganzha1 aLeszek Maciaszek1 aMarcin Paprzycki1 aDominik Slezak00aContinual learning of a time series model using a mixture of HMMs with application to the IoT fuel sensor verification uhttps://doi.org/10.15439/2023F185600754nas a2200253 4500000000100000000000100001008004100002260001200043100002000055700002200075700002300097700002400120700001800144700001900162700001900181700001700200700001900217700001900236700002100255245013000276856007200406490000800478022001400486 2024 d c05/20241 aKamil Książek1 aWilhelm Masarczyk1 aPrzemysław Głomb1 aMichał Romaszewski1 aIga Stokłosa1 aPiotr Ścisło1 aPaweł Dębski1 aRobert Pudlo1 aKrisztian Buza1 aPiotr Gorczyca1 aMagdalena Piegza00aAssessment of symptom severity in psychotic disorder patients based on heart rate variability and accelerometer mobility data uhttps://www.sciencedirect.com/science/article/pii/S00104825240062800 v176 a0010-482500457nas a2200157 4500000000100000000000100001008004100002260001200043100002400055700002400079700002300103700002000126700002500146245009100171856003700262 2024 d c06/20241 aMichał Romaszewski1 aPrzemysław Sekuła1 aPrzemysław Głomb1 aMichał Cholewa1 aKatarzyna Kołodziej00aThrough the Thicket: A Study of Number-Oriented LLMs derived from Random Forest Models uhttps://arxiv.org/abs/2406.0492600578nas a2200193 4500000000100000000000100001008004100002260001200043100002000055700002400075700002300099700002500122700002100147700001600168700001900184700001800203700002000221245014300241 2024 d c06/20241 aMichał Cholewa1 aMichał Romaszewski1 aPrzemysław Głomb1 aKatarzyna Kołodziej1 aMichał Gorawski1 aJakub Koral1 aWojciech Koral1 aAndrzej Madej1 aKryspin Musioł00a‘Just One More Sensor is Enough’ – Iterative Water Leak Localization with Physical Simulation and a Small Number of Pressure Sensors00456nas a2200157 4500000000100000000000100001008004100002260001200043100002500055700002000080700002300100700001900123700002400142245009500166856003700261 2024 d c09/20241 aKatarzyna Kołodziej1 aMichał Cholewa1 aPrzemysław Głomb1 aWojciech Koral1 aMichał Romaszewski00aEfficient Numerical Calibration of Water Delivery Network Using Short-Burst Hydrant Trials uhttps://arxiv.org/abs/2410.0277200605nas a2200217 4500000000100000000000100001008004100002100001700043700002500060700002300085700002300108700002500131700002000156700002100176700002200197700002400219700001800243700002200261700001900283245008500302 2024 d1 aPiotr Gawron1 aPrzemysław Sadowski1 aPrzemysław Głomb1 aBartłomiej Gardas1 aMatthijs Van Waveren1 aClément Forray1 aGuillaume Pasero1 aMickaël Savinaud1 aPierre-Marie Brunet1 aOrphee Faucoz1 aZbigniew Puchała1 aŁukasz Pawela00aWhat Could be Achieved with a Million Qubits Quantum Annealer in Remote Sensing?00409nas a2200145 4500000000100000000000100001008004100002100002500043700002000068700002400088700002300112700001900135700002400154245008500178 2024 d1 aKatarzyna Kołodziej1 aMichał Cholewa1 aMichał Romaszewski1 aPrzemysław Głomb1 aWojciech Koral1 aPrzemysław Sekuła00aCalibration of Water Demand Network Through a Machine Learned Digital Twin Model00427nas a2200157 4500000000100000000000100001008004100002260001200043100002400055700002400079700002300103700002000126700002500146245009100171490000700262 2025 d c03/20251 aMichał Romaszewski1 aPrzemysław Sekuła1 aPrzemysław Głomb1 aMichał Cholewa1 aKatarzyna Kołodziej00aThrough the Thicket: A Study of Number-Oriented LLMS Derived from Random Forest Models0 v1501940nas a2200157 4500000000100000000000100001008004100002100002400043700002400067700002300091700002000114700001900134245006100153856006700214520150100281 2025 d1 aPrzemysław Sekuła1 aMichał Romaszewski1 aPrzemysław Głomb1 aMichał Cholewa1 aŁukasz Pawela00aQuantum-aware Transformer model for state classification uhttps://link.springer.com/chapter/10.1007/978-3-031-97570-7_153 a

Entanglement is a fundamental feature of quantum mechanics, playing a crucial role in quantum information processing. However, classifying entangled states, particularly in the mixed-state regime, remains a challenging problem, especially as system dimensions increase. In this work, we focus on bipartite quantum states and present a data-driven approach to entanglement classification using transformer-based neural networks. Our dataset consists of a diverse set of bipartite states, including pure separable states, Werner entangled states, general entangled states, and maximally entangled states. We pretrain the transformer in an unsupervised fashion by masking elements of vectorized Hermitian matrix representations of quantum states, allowing the model to learn structural properties of quantum density matrices. This approach enables the model to generalize entanglement characteristics across different classes of states. Once trained, our method achieves near-perfect classification accuracy, effectively distinguishing between separable and entangled states. Compared to previous Machine Learning, our method successfully adapts transformers for quantum state analysis, demonstrating their ability to systematically identify entanglement in bipartite systems. These results highlight the potential of modern machine learning techniques in automating entanglement detection and classification, bridging the gap between quantum information theory and artificial intelligence.

02630nas a2200241 4500000000100000000000100001008004100002260001200043100002000055700002200075700002300097700002400120700001900144700002400163700002000187700002500207700001900232700002100251245012200272490000700394520197300401022001402374 2025 d c09/20251 aKamil Książek1 aWilhelm Masarczyk1 aPrzemysław Głomb1 aMichał Romaszewski1 aKrisztian Buza1 aPrzemysław Sekuła1 aMichał Cholewa1 aKatarzyna Kołodziej1 aPiotr Gorczyca1 aMagdalena Piegza00aDeep learning approach for automatic assessment of schizophrenia and bipolar disorder in patients using R-R intervals0 v213 a

Schizophrenia and bipolar disorder are severe mental illnesses that significantly impact quality of life. These disorders are associated with autonomic nervous system dysfunction, which can be assessed through heart activity analysis. Heart rate variability (HRV) has shown promise as a potential biomarker for diagnostic support and early screening of those conditions. This study aims to develop and evaluate an automated classification method for schizophrenia and bipolar disorder using short-duration electrocardiogram (ECG) signals recorded with a low-cost wearable device. We conducted classification experiments using machine learning techniques to analyze R-R interval windows extracted from short ECG recordings. The study included 60 participants—30 individuals diagnosed with schizophrenia or bipolar disorder and 30 control subjects. We evaluated multiple machine learning models, including Support Vector Machines, XGBoost, multilayer perceptrons, Gated Recurrent Units, and ensemble methods. Two time window lengths (about 1 and 5 minutes) were evaluated. Performance was assessed using 5-fold cross-validation and leave-one-out cross-validation, with hyperparameter optimization and patient-level classification based on individual window decisions. Our method achieved classification accuracy of 83% for the 5-fold cross-validation and 80% for the leave-one-out scenario. Despite the complexity of our scenario, which mirrors real-world clinical settings, the proposed approach yielded performance comparable to advanced diagnostic methods reported in the literature. The results highlight the potential of short-duration HRV analysis as a cost-effective and accessible tool for aiding in the diagnosis of schizophrenia and bipolar disorder. Our findings support the feasibility of using wearable ECG devices and machine learning-based classification for psychiatric screening, paving the way for further research and clinical applications.

 a1553-735800311nas a2200109 4500000000100000000000100001008004100002100002500043700002300068700001800091245009200109 2025 d1 aKatarzyna Kołodziej1 aPrzemysław Głomb1 aAnna Zawadzka00aChallenges associated with the integration of Large Multimodal Models for waste sorting00317nas a2200109 4500000000100000000000100001008004100002100002000043700002400063700002300087245009700110 2025 d1 aMichał Cholewa1 aMichał Romaszewski1 aPrzemysław Głomb00aData structure better than labels? Unsupervised heuristics for SVM hyperparameter estimation00317nas a2200109 4500000000100000000000100001008004100002100002000043700002400063700002300087245009700110 2025 d1 aMichał Cholewa1 aMichał Romaszewski1 aPrzemysław Głomb00aData structure better than labels? Unsupervised heuristics for SVM hyperparameter estimation01850nas a2200121 4500000000100000000000100001008004100002100001800043700002300061700002000084245008900104520153500193 2025 d1 aAnna Zawadzka1 aPrzemysław Głomb1 aMateusz Żarski00aDevelopment of an effective method for detecting small buildings in satellite images3 a

Detecting and monitoring small-scale buildings is a critical challenge in various sectors and can be crucial for various important tasks. Buildings considered small are those with a footprint smaller than 10 × 10 meters, a definition justified by the resolution of Sentinel-2 images, which is 10 meters at best. Examples of tasks for which small building detection is essential include assessment and monitoring of damage caused by natural disasters or war, identification of illegal settlements, detecting new buildings in key military areas, and identification of structures in protected areas. Regrettably, widely available satellite imagery, such as Sentinel-2, provides very limited spatial resolution, making it impossible to efficiently detect small structures. Despite its importance, this issue remains unexplored, with very limited research directly concerning building segmentation and detecting small structures. Traditional segmentation and classification methods exhibit severe limitations, particularly due to small buildings frequently being positioned at pixel boundaries, essentially occupying an insignificant fraction of the original pixel. Additionally, in the context of buildings in complex environments (e.g., surrounded by vegetation or shadows), popular methods such as U-Net with ResNet often fail to deliver reliable results. In this study, we propose a novel approach to this issue by integrating a state-of-the-art object detection model with a deep-learning-based super-resolution network.

00485nas a2200169 4500000000100000000000100001008004100002260001200043100002500055700002000080700002300100700001900123700002400142245009500166856004600261490000800307 2026 d c02/20261 aKatarzyna Kołodziej1 aMichał Cholewa1 aPrzemysław Głomb1 aWojciech Koral1 aMichał Romaszewski00aEfficient Numerical Calibration of Water Delivery Network Using Short-Burst Hydrant Trials uhttps://doi.org/10.1061/JWRMD5.WRENG-68390 v15200365nas a2200145 4500000000100000000000100001008004100002100001800043700002000061700001700081700002300098700002400121700002000145245005400165 2026 d1 aAnna Zawadzka1 aMateusz Żarski1 aKamil Drejer1 aPrzemysław Głomb1 aMichał Romaszewski1 aMichał Cholewa00a Map-Guided Cross-Training for Building Detection01085nas a2200109 4500000000100000008004100001260001200042100001800054700002300072245007500095520080500170 2026 d c04/20261 aAnna Zawadzka1 aPrzemysław Głomb00aArchitecture Matters: Gender Disparities in Automated Image Moderation3 a

Automated image moderation systems shape online visibility and dataset curation, yet prior work has identified demographic disparities in similarity-based NSFW classifiers. We compare such systems with instruction-tuned vision-language models (VLMs) that generate structured moderation decisions with textual rationales. Using the PHASE-annotated subset of the GCC dataset,  we compute false positive rates overall and by gender. 
Results show substantial variation in moderation strictness and in bias direction: two models show more pronounced strictness in moderation of male images, while other two exhibit higher removal rates for female images. 
Reasoning-based models do not consistently mitigate bias but, in the proposed pipeline, they offer a transparency layer.

02216nas a2200169 4500000000100000008004100001100001700042700002500059700002000084700002400104700002300128700001800151700001600169700001900185245007600204520176600280 2026 d1 aAygün Varol1 aKatarzyna Kołodziej1 aŁukasz Sobczak1 aMichał Romaszewski1 aPrzemysław Głomb1 aNaser Motlagh1 aMirka Leino1 aJohanna Virkki00aEnabling Cloud-Level Accuracy in Edge AI through IoT Data Preprocessing3 a

Large language models (LLMs) offer a natural-language interface for interpreting Internet of Things (IoT) sensor data in smart environments; however, cloud deployment introduces latency, privacy, and connectivity concerns. Local LLMs can reduce these limitations, but compact edge-deployable models often show weaker numerical reasoning when raw sensor readings are provided directly. This paper investigates whether prompt-side preprocessing can improve the accuracy--latency trade-off of local LLMs for environmental monitoring. We propose a structured prompt construction framework that transforms raw air-quality and thermal-comfort measurements into progressively enriched textual representations: raw sensor values, threshold-aware descriptions, and compact environmental summary flags. The approach is evaluated using indoor Raspberry Pi/BME680 datasets from Tampere University and outdoor air-quality datasets from Helsinki, Katowice, and Warsaw. We construct a binary LLM query dataset covering air quality, thermal comfort, and joint environmental conditions, and evaluate five local and five cloud LLMs across three prompt variants and two inference modes, with and without chain-of-thought prompting. The results show that prompt enrichment substantially improves local-model accuracy. In No-CoT mode, local accuracy increases from 50.9\% to 81.7\% indoors and from 63.7\% to 89.3\% outdoors from the raw to the most enriched prompt. Local No-CoT inference is the fastest configuration, with mean latency close to 0.22 s, while CoT substantially increases inference time. These findings suggest that lightweight prompt-side preprocessing can narrow the local--cloud performance gap and support low-latency IoT analytics in smart environments.

02266nas a2200181 4500000000100000008004100001100001700042700002500059700002000084700002400104700002300128700001800151700001600169700001900185245007600204856003800280520176600318 2026 d1 aAygün Varol1 aKatarzyna Kołodziej1 aŁukasz Sobczak1 aMichał Romaszewski1 aPrzemysław Głomb1 aNaser Motlagh1 aMirka Leino1 aJohanna Virkki00aEnabling Cloud-Level Accuracy in Edge AI through IoT Data Preprocessing uhttps://arxiv.org/abs/2606.22496 3 a

Large language models (LLMs) offer a natural-language interface for interpreting Internet of Things (IoT) sensor data in smart environments; however, cloud deployment introduces latency, privacy, and connectivity concerns. Local LLMs can reduce these limitations, but compact edge-deployable models often show weaker numerical reasoning when raw sensor readings are provided directly. This paper investigates whether prompt-side preprocessing can improve the accuracy--latency trade-off of local LLMs for environmental monitoring. We propose a structured prompt construction framework that transforms raw air-quality and thermal-comfort measurements into progressively enriched textual representations: raw sensor values, threshold-aware descriptions, and compact environmental summary flags. The approach is evaluated using indoor Raspberry Pi/BME680 datasets from Tampere University and outdoor air-quality datasets from Helsinki, Katowice, and Warsaw. We construct a binary LLM query dataset covering air quality, thermal comfort, and joint environmental conditions, and evaluate five local and five cloud LLMs across three prompt variants and two inference modes, with and without chain-of-thought prompting. The results show that prompt enrichment substantially improves local-model accuracy. In No-CoT mode, local accuracy increases from 50.9\% to 81.7\% indoors and from 63.7\% to 89.3\% outdoors from the raw to the most enriched prompt. Local No-CoT inference is the fastest configuration, with mean latency close to 0.22 s, while CoT substantially increases inference time. These findings suggest that lightweight prompt-side preprocessing can narrow the local--cloud performance gap and support low-latency IoT analytics in smart environments.

03010nas a2200181 4500000000100000008004100001100001700042700002500059700002000084700002400104700002300128700001800151700001600169700001900185245011600204856005200320520245600372 2026 d1 aAygün Varol1 aKatarzyna Kołodziej1 aŁukasz Sobczak1 aMichał Romaszewski1 aPrzemysław Głomb1 aNaser Motlagh1 aMirka Leino1 aJohanna Virkki00aAIRWISE: Environmental Sensor and LLM Query Dataset for Edge AI-Based Air-Quality and Thermal-Comfort Analytics uhttps://zenodo.org/doi/10.5281/zenodo.20783874 3 a

AIRWISE is an open benchmark dataset for edge AI, IoT environmental monitoring, and large language model (LLM) evaluation on real sensor data. It combines indoor air-quality and thermal-comfort measurements, outdoor air-quality and meteorological time series, and a labeled binary question-answering benchmark for threshold-aware reasoning and anomaly detection. The dataset supports the study “Enabling Cloud-Level Accuracy in Edge AI through IoT Data Preprocessing”: https://doi.org/10.48550/arXiv.2606.22496
AIRWISE provides three core resources:
Indoor environmental sensor dataset (BME680, 1-minute resolution). 146,000+ minute-averaged readings from Raspberry Pi + Bosch BME680 nodes in three real micro-environments (office, kitchen, hallway) at Tampere, Finland (2025-11-14 to 2025-12-18): temperature, relative humidity, barometric pressure, gas resistance, indoor air quality (IAQ) proxy, z-score anomaly detectors, and limit-annotated long-format tables with ground-truth anomaly flags.
Outdoor air-quality and meteorological time series (hourly, full-year 2023). Helsinki (Finland), Katowice and Warsaw (Poland): NO₂, CO, O₃, PM₂.₅, temperature, and relative humidity assembled from national monitoring networks (FMI, GIOŚ, IMGW-PIB), as cleaned single-series city tables (≈8,760 hourly records per city) and limit-annotated long-format tables with anomaly flags.
Binary LLM query benchmark (1,440 labeled yes/no questions). 240 natural-language questions per site across six sites, balanced by task type (80 air quality, 80 thermal comfort, 80 joint environmental condition), each with ground-truth answer, violation label, offending factors, and rationale — ready for benchmarking local/edge and cloud LLMs on sensor-grounded reasoning.
Also included: reference air-quality and thermal-comfort limits (EU air-quality standards and Finnish/Polish national guidelines); the unmodified institutional source files for the outdoor component (FMI exports, GIOŚ archive workbooks and station metadata, IMGW-PIB synoptic tables); the complete dataset-generation scripts covering every pipeline step (raw sources → cleaned tables → limit-annotated data → LLM queries) with a requirements.txt for a verified minimal Python environment; SHA-256 checksums; and a dependency-free validation script. Schemas, exact row counts, coverage, and usage examples are documented in README.md and MANIFEST.md.

00582nas a2200193 4500000000100000008004100001100002400042700001700066700002500083700001800108700002500126700002300151700002000174700002300194700002100217700002400238700001600262245011000278 2026 d1 aMichał Romaszewski1 aKamil Drejer1 aKatarzyna Kołodziej1 aAnna Zawadzka1 aStanisław Lewiński1 aPrzemysław Głomb1 aMarek Ruciński1 aMichał Krupiński1 aKrzysztof Gryguc1 aPrzemysław Sekuła1 aSzymon Sala00aToward Seasonal Guidelines for Robust Deep-Learning Sentinel-2 Building Detection in Different Area Types