Image Processing and Pattern Recognition
Web site: http://www.cv.utcluj.roResearch team
Sergiu Nedevschi, Tiberiu Marita, Radu Danescu, Florin Oniga, Ciprian Pocol, Delia Mitrea, Pangyu Jeong, Stefan Sobol, Corneliu Tomiuc, Cristian Vancea, Silviu Bota, Raluca Borca, Mihai Negru, Cristian Vicas, Andrei Vatavu, Ioan Giosan, Radu Lazar, Cosmin Pantilie
Research fields
- 2D and 3D Image Processing and Recognition
- Features extraction and measurements from intensity images
- Pattern recognition
- Image texture detection and analysis
- Camera calibration
- Binocular and trinocular stereovision
- Stereovision for mobile robots and automotive applications
- Design and implementation of hardware solutions for stereovision using FPGA technology
- Stereovision for Driving Assistance Systems and Autonomous Driving
- 3D lane detection
- 3D objects detection and tracking
- 3D objects classification
- Pedestrian detection
- 3D structured/unstructured environment modeling
- Medical Image Processing
- Texture based detection and classification of diffuse and focal illnesses from ultrasonography
- Structured reporting of medical images
- DICOM infrastructure implementation
Recent Research Projects
Stereovision and driving assistance:
“STEREOSENS - Stereo Sensor for City Automatic Cruise Control” research project funded by Volkswagen AG, Germany, (2007).
S. Nedevschi coordinator, R. Danescu scientific coordinator, T. Marita, F. Oniga, C. Pocol, S. Sobol, S. Bota, C. Tomiuc, V. Cristian, A. Vatavu, M. Negru,
“STEREOPED - Dense Stereo-Based Object Tracking and Pedestrian Recognition for Pre-Crash-Applications” Germany, (2007). research project funded by Volkswagen AG,
S. Nedevschi coordinator, S. Bota scientific coordinator, T. Marita, R. Danescu, F. Oniga, C. Pocol, S. Sobol, C. Tomiuc, V. Cristian, R. Lazar, I. Giosan
"DESPED - Stereo-Based Object Tracking and Pedestrian Recognition in Traffic environments" research project funded by Volkswagen AG, Germany, (2006-2007).
S. Nedevschi coordinator, T. Marita scientific coordinator, Dorian Popa, R. Danescu, F. Oniga, C. Pocol, S. Sobol, C. Tomiuc, V. Cristian, S. Bota, R. Lazar, R. Borca M. Negru, I. Giosan.
"DESBOR - Dense Stereo-Based Object Recognition system for automatic cruise control (ACC) in urban traffic environments" research project funded by Volkswagen AG, Germany, (2005-2007).
S. Nedevschi coordinator, F. Oniga scientific coordinator, T. Marita, R. Danescu, C. Pocol, S. Sobol, S. Bota, C. Tomiuc, V. Cristian, R. Borca, M. Negru, A. Vatavu, R. Lazar, D. Popa
"SCABOR - Stereo Camera Based Object Recognition System for Vehicle Application" research project funded by Volkswagen AG, Germany, (2004)
S. Nedevschi coordinator, T. Marita scientific coordinator, R. Danescu, F. Oniga, C. Pocol, S. Sobol, C. Tomiuc, C, Vancea.
"TRINOCULAR - Method and system for real-time acquisition of high-resolution 3D images based on trinocular stereo-vision" national research grant funded by CNCSIS, A – 884, (2004 - 2006).
S. Nedevschi coordinator, S. Bota, T. Marita, R. Danescu, F. Oniga, C. Pocol, S. Sobol, C. Tomiuc, C. Vancea, R. Lazar, R. Borca, M. Negru, A. Vatavu.
Medical imaging:
"Intelligent System for the Noninvasive Detection and Evaluation of Liver Fibrosis, Restructuration and Dysplastic Nodules, Using 2d/3D Ultrasound and Molecular Markers - SIDEF", national research grant funded by ANCS, CEEX VIASAN 71/2006, (2006 - 2008).
S. Nedevschi scientific coordinator, I. Salomie, R. Potolea, M. Dinsoreanu, T. Marita, P. Mitrea, R. Danescu, F. Oniga, C. Pocol, D. Mitrea, S. Stefan, C. Tomiuc, C. Vancea, C. Vicas, I. Giosan.
“Non-alcoholic fatty liver, viral C hepatitis and gallstone disease – components of the metabolic syndrome. Clinical epidemiology, pathogenesis, non-invasive diagnosis – FINALISM”, national research grant funded by ANCS, CEEX VIASAN 94/2006, (2006-2008).
S. Nedevschi scientific coordinator, T. Marita, R. Danescu, F. Oniga, C. Pocol, D. Mitrea, C. Vicas.
“Laparoscopic cryosurgical treatment of the renal tumors, individualized using simulation on three-dimensional reconstructed model – CRIOLAPSIM”, national research grant funded by ANCS, CEEX VIASAN 121/2006, (2006-2008).
S. Nedevschi scientific coordinator, M. Ivan, I. Rasa, D. Popa, T. Marita, M. Gurzau, I. Matis, A. Vatavu, R. Borca, C. Pantilie, M. Moldovan, A. Faur.
“Molecular diagnosis of epidermolysis bullosa. Modern techniques for research, diagnosis, treatment and prevention of epidermolysis bullosa. Implementation of a national record of genodermatoses – DEGEB”, national research grant funded by ANCS, CEEX VIASAN 126/2006, (2006-2008).
S. Nedevschi scientific coordinator, R. Danescu, C. Lelutiu, C. Tomiuc, S. Sobol, C. Vancea, A. Vatavu
"Intelligent System for assisting the therapeutical decision at patients with prostate cancer - INTELPRO", national research grant funded by ANCS, CEEX-INFOSOC, 18/2005 (2005 - 2008).
S. Nedevschi coordinator, M. Dinsoreanu scientific coordinator, I. Salomie, R. Potolea, T. Muresan, T. Marita, C. Cenan, D. Mitrea, C. Vidrighin, et al.
"Telemedicine Network for Ultrasound Screening and Monitoring of Hepatocellular Carcinoma – TELEHEPASCAN", national research grant funded by ANCS, CEEX VIASAN 3/2005, (2005 - 2008).
S. Nedevschi scientific coordinator, T. Marita, R. Danescu, F. Oniga, C. Pocol, D. Mitrea, C. Vicas, M. Rus.
Publications
Stereo vision and driving assistance:
S. Nedevschi, C. Vancea, T. Marita, T. Graf, “On-Line Calibration Method for Stereovision Systems Used in Far Range Detection Vehicle Applications”, IEEE Transactions on Intelligent Transportation Systems, vol. 8, no. 4, December, 2007, pp. 651-660.
R. Danescu, S. Nedevschi, M.M. Meinecke, T.B. To, “Lane Geometry Estimation in Urban Environments Using a Stereovision System”, Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC 2007), Seattle, USA, September 30 – October 3, 2007, pp. 271-276.
F. Oniga, S. Nedevschi, M. M. Meinecke, and T. B. To, “Road Surface and Obstacle Detection Based on Elevation Maps from Dense Stereo”, Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC 2007), Seattle, USA, September 30 – October 3, 2007, pp. 401-404.
R. Danescu, S. Nedevschi, M. M. Meinecke, T. Graf, “Stereovision Based Vehicle Tracking in Urban Traffic Environments”, Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC 2007), Seattle, USA, September 30 – October 3, 2007, pp. 859-865.
S. Nedevschi, R. Danescu, T. Marita, F. Oniga, C. Pocol, S. Sobol, C. Tomiuc, C. Vancea, M. M. Meinecke, T. Graf, T. B. To, M. A. Obojski, “A Sensor for Urban Driving Assistance Systems Based on Dense Stereovision”, Proceedings of IEEE International Symposium Intelligent Vehicles (IV 2007), Istambul, June 13-15, 2007, pp.278-286.
S. Nedevschi, C. Tomiuc, S. Bota, "Stereo Based Pedestrian Detection for Collision Avoidance Applications", Proceedings of ICRA 2007 Workshop: Planning, Perception and Navigation for Intelligent Vehicles, Roma, April, 2007, pp. 39-44.
S. Bota, S. Nedevschi, "Multi-Feature Walking Pedestrian Detection Using Dense Stereo Motion", Proceedings of WIT 2007, Hamburg, 20-21 March, 2007, pp. 113-118.
P. Jeong, S. Nedevschi, “Local Difference Probability (LDP)-Based Environment Adaptive Algorithm for Unmanned Ground Vehicle”, IEEE Transactions on Intelligent Transportation Systems, vol. 7, no. 3, pp. 282 – 292, September, 2006.
S. Nedevschi, C. Vancea, T. Marita, T. Graf, "On-Line Calibration Method for Stereovision Systems Used in Vehicle Applications”, Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC 2006), Toronto, Canada, September 17-20, 2006, pp. 957-962.
R. Danescu, S. Sobol, S. Nedevschi, T. Graf, "Stereovision-Based Side Lane and Guardrail Detection", Proceedings of the IEEE Intelligent Transportation Systems Conference (ITSC 2006), Toronto, Canada, September 17-20, 2006, pp. 1156-1161.
S. Nedevschi, F. Oniga, R. Danescu, T. Graf, R. Schmidt, "Increased Accuracy Stereo Approach for 3D Lane Detection", Proceedings of IEEE Intelligent Vehicles Symposium (IV 2006), Tokyo, Japan, June 13-15, 2006, pp. 42-49.
T. Marita, F. Oniga, S. Nedevschi, T. Graf, R. Schmidt, "Camera Calibration Method for Far Range Stereovision Sensors Used in Vehicles", Proceedings of
IEEE Intelligent Vehicles Symposium (IV 2006), Tokyo, Japan, June 13-15, 2006, pp. 356-363.
C. Vancea, S. Nedevschi, M. Negru, St. Mathe, “Real-Time FPGA-based Image Rectification System”, Proceedings of First International Conference on Computer Vision Theory and Applications (VISAPP 2006), Setubal, Portugalia, February 25-28, 2006, vol. 1, pp. 93-100.
S. Nedevschi, R. Danescu,T. Marita, F. Oniga, C. Pocol, S. Sobol, T. Graf, R. Schmidt, "Driving Environment Perception Using Stereovision", Procedeeings of IEEE Intelligent Vehicles Symposium (IV 2005), Las Vegas, USA, June 2005, pp.331-336.
S. Nedevschi, R. Schmidt, T. Graf, R. Danescu, D. Frentiu, T. Marita, F. Oniga, C. Pocol, “3D Lane Detection System Based on Stereovision”, Proceedings of IEEE Intelligent Transportation Systems Conference (ITSC’04), Washington, USA, 4-6 October, 2004, pp. 161-166.
S. Nedevschi, S. Mathe, “FPGA-Based Edge Detection with Subpixel Accuracy”, Proceedings of IEEE Intelligent Engineering Systems (INES), September 2004, Cluj Napoca, Romania, pp. 447-452.
D. Mitrea, S. Nedevschi, “Road quality evaluation and road material recognition using 3D textons”, Proceedings of 8th IEEE International Conference on Intelligent Engineering Systems (INES), Cluj-Napoca, Romania, September, 19-21, pp. 236-241.
S. Nedevschi, R. Danescu, D. Frentiu, T. Marita, F. Oniga, C. Pocol, R. Schmidt, T. Graf, “High accuracy stereo vision system for far distance obstacle detection”, Proceedings of IEEE Intelligent Vehicles Symposium, Parma (IV 2004), Italy, June 14-17, 2004, pp. 292-297.
S. Nedevschi, R. Danescu, D. Frentiu, T. Marita, F. Oniga, C. Pocol, “Spatial Grouping of 3D Points from Multiple Stereovision Sensors”, Proceedings of IEEE International Conference on Networking, Sensing and Control, Taipei, 21-23 March, 2004, pp. 874-879.
Medical imaging:
D. Mitrea, S. Nedevschi, M. Lupsor, R. Badea, I. Coman, “Exploring the textural parameters of ultrasound images to build an imagistic model for prostatic adenocarcinoma (ADKP)”, Workshop on Computers in Medical Diagnoses WCMD 2007, Cluj-Napoca, Romania, September 6-8, 2007, published in Automation
Computers Applied Mathematics. Scientific Journal (ACAM), vol. 16, no. 1, pp. 11-19, 2007.
C. Vicas, S. Nedevschi, M. Lupsor, R. Badea, H. Stefanescu, "Fibrosis detection from ultrasound imaging. The influence of necro-inflammatory activity and steatosis over the detection rates", Workshop on Computers in Medical Diagnoses WCMD 2007, Cluj-Napoca, Romania, September 6-7, 2007, published in Automation Computers Applied Mathematics. Scientific Journal (ACAM), vol. 16, no. 1, pp. 27-33, 2007
C. Vicas, S. Nedevschi, M. Lupsor, R. Badea, M. Grigorescu, "Steatohepatitis Detection from Ultrasound Images Using Attenuation and Backscattering Coefficients", Workshop on Computers in Medical Diagnoses WCMD 2007, Cluj-Napoca, Romania, September 6-7, 2007, published in Automation Computers Applied Mathematics. Scientific Journal (ACAM), vol. 16, no. 1, pp. 20-26, 2007
S. Dudea, S. Nedevschi, C. Pantilie, C. Botar-Jid, D. Dumitriu, T. Marita, “Ultrasound Elastography: From Phisical Pricipals to Computer Aided Image Analysis and Quantification”, Workshop on Computers in Medical Diagnoses WCMD 2007, Cluj-Napoca, Romania, September 6-7, 2007, published in ACAM Scientific Journal, vol. 16, no. 1, pp. 34-39, 2007.
D. Mitrea, S. Nedevschi, P. Mitrea, M. Lupsor, R. Badea, M. Grigorescu, S. Tripon, C. Radu, “Texture-based Methods and Data-Mining in Ultrasound Images for Diffuse Liver Disease Severity Characterization”, Advancements of Medicine and Health care through Technology (MEDITECH 2007), Cluj-Napoca, Romania, September 2007, published in Acta Electrotehnica (Special Issue), vol. 48, no. 4, pp. 211-216, 2007.
D. Mitrea, S. Nedevschi, C. Cenan, M. Lupsor, R. Badea, “Exploring Texture-Based Parameters, Noninvasive Characterization and Modelling of Diffuse Liver Diseases and Liver Cancer from Ultrasound Images”, WSEAS Transactions on Computers, vol. 6, no. 2, pp. 283-291, February 2007.
C. Homorodean, D. Olinic, S. Nedevschi, N. Olinic, “Templates implementation for structured DICOM diagnosis reporting in echocardiography”, Proceedings of 2005 IEEE International Conference Computers in Cardiology, September 25-28 2005, Lyon, France, pp. 379 – 382.
S. Nedevschi, F Rusu, I. Popa, A. Smeu, D. Olinic, C. Homorodean, “Structured Reporting of Echocardiography Images in a DICOM Environment”, Proceedings of 2004 IEEE International Conference Computers in Cardiology, Chicago, September 19-22, 2004, pp. 685-688.
D. Olinic, C. Homorodean, S. Nedevschi, F. Rusu, A. Smeu, I. Popa, A. Vancea, N. Olinic, “A proposal for structured diagnosis reporting in echocardiography, using a DICOM compliant environment”, Proceedings of 2004 IEEE International Conference Computers in Cardiology, Chicago, September 19-22, 2004, pp. 601-604.
Research description
Dense Stereo-Based Object Recognition system for automatic cruise control (ACC) in urban traffic environments (DESBOR)
Motivation
Automotive companies are investing a lot of interest in intelligent systems which can assist the driving process. Radar and laser sensors based applications are dedicated for precise pose measurements but they cannot give a complete description of the driving environment due to their nature. Vision sensors can compensate the information lack of their measurements. Stereovision has been proved the only method able give reliable results both for measurements and description of the driving environment.
Objectives
Development of a real time dense stereo camera based system able to detect, classify and track relevant objects, detect the lane and estimate the road geometry, detect and recognize traffic signs painted on the roads. Based on the detection parameters the system must be able to perform specific urban traffic ACC tasks: stop & go (follow to stop and stop-inhibit), vehicle lateral control, set maximum ACC velocity etc.
Marginal conditions:
- Detection range: 0 to 50 m
- Field of view: 72 degrees
- Maximum traveling speed: 80 km/h
- Processing rate: 15 frames/secon
Project activities
Dense stereovision
Realtime dense stereo reconstruction is performed using a dedicated board (DeepSea by TYZX). A software algorithm was also developed for offline validation purposes.
Detection of elevation maps
An original and robust solution for lane detection using 3D road features, a model driven approach and Kalman filter tracking was developed (Fig.2). It uses non-flat road assumption, and works even in extreme conditions (absence of lane markings - Fig. 3). Side lanes and driving area (with delimiters) are also detected (Fig. 2). Detected lane parameters are: vertical and horizontal curvature, lane width, and lateral position of the ego-car inside the current lane.
Object detection and tracking
The equation of the road surface is used for obstacle/road point separation and therefore object detection works even on non-flat road scenarios. The overrode 3D points are grouped into objects taking into account the vicinity criteria. The 3D density variation with the distance is taken into account. 2D image information such as similar texture and connecting edges are refining the grouping process. The object's position is tracked over successive frames using a mathematical motion model and Kalman filtering. Detected objects are described in terms of 3D position, size and speed.
Scientific achievements
- Fast automatic gain and exposure control using a non-linear camera lighting response mode
- Camera calibration by fast nonlinear optimization of the camer parameter set
- Template matching for calibration target recognition at sub-pixel level calibration methodology automation
- Real-time image scaling and rectification using a distortion-aware camera model for dense stereovision
- Model-based road surface estimation from dense stereo data using the RANSAC approach
- Freeform road surface estimation using local constraints
- Model-based lane estimation using 3D points and image data under a probabilistic estimation framework
- Probabilistic fusion of heterogeneous lane cues
- Lane tracking using the vehicle's motion parameters
- Object detection by clustering of 3D points of variable density into 3D cuboids
- Probabilistic cuboid tracking using the Kalman filter and the vehicle motion laws
Products delivered by the research activity
Dense stereovision-based system for relevant object detection and measurement in urban traffic scenarios, for automatic cruise control. The system is composed of the hardware architecture (image acquisition and synchronization setup, processing platform, car network interface and standard car motion sensors) and the software product, capable of processing the images and the sensor data and delivering the detection results.