National Research Council Canada Conseil national de recherches Canada National Research Council Canada Conseil national de recherches Canada Institute.

Slides:



Advertisements
Similar presentations
Evidential modeling for pose estimation Fabio Cuzzolin, Ruggero Frezza Computer Science Department UCLA.
Advertisements

National Research Council Canada Conseil national de recherches Canada National Research Council Canada Conseil national de recherches Canada Canada Dmitry.
By: Mani Baghaei Fard.  During recent years number of moving vehicles in roads and highways has been considerably increased.
QR Code Recognition Based On Image Processing
Joshua Fabian Tyler Young James C. Peyton Jones Garrett M. Clayton Integrating the Microsoft Kinect With Simulink: Real-Time Object Tracking Example (
Image Analysis Phases Image pre-processing –Noise suppression, linear and non-linear filters, deconvolution, etc. Image segmentation –Detection of objects.
System Integration and Experimental Results Intelligent Robotics Research Centre (IRRC) Department of Electrical and Computer Systems Engineering Monash.
Lecture 07 Segmentation Lecture 07 Segmentation Mata kuliah: T Computer Vision Tahun: 2010.
Where has all the data gone? In a complex system such as Metalman, the interaction of various components can generate unwanted dynamics such as dead time.
‘ Glaucoma Detection In Retinal Images Using Automated Method ’
GIS and Image Processing for Environmental Analysis with Outdoor Mobile Robots School of Electrical & Electronic Engineering Queen’s University Belfast.
Uncertainty Representation. Gaussian Distribution variance Standard deviation.
DTM Generation From Analogue Maps By Varshosaz. 2 Using cartographic data sources Data digitised mainly from contour maps Digitising contours leads to.
INTRODUCTION. Painting with numbers! Aspects Modeling Rendering Animation.
Applications of Image Processing. Outline What YOU are going to learn Image Processing in use.
Vision-Based Motion Control of Robots
January 19, 2006Computer Vision © 2006 Davi GeigerLecture 1.1 Image Measurements and Detection Davi Geiger
A Study of Approaches for Object Recognition
Vision Computing An Introduction. Visual Perception Sight is our most impressive sense. It gives us, without conscious effort, detailed information about.
Highlights Lecture on the image part (10) Automatic Perception 16
Scale Invariant Feature Transform (SIFT)
Image Processing of Video on Unmanned Aircraft Video processing on-board Unmanned Aircraft Aims to develop image acquisition, processing and transmission.
Inverse Kinematics Problem: Input: the desired position and orientation of the tool Output: the set of joints parameters.
Shadow Removal Seminar
CS292 Computational Vision and Language Visual Features - Colour and Texture.
Jason Li Jeremy Fowers Ground Target Following for Unmanned Aerial Vehicles.
1 REAL-TIME IMAGE PROCESSING APPROACH TO MEASURE TRAFFIC QUEUE PARAMETERS. M. Fathy and M.Y. Siyal Conference 1995: Image Processing And Its Applications.
A Brief Overview of Computer Vision Jinxiang Chai.
A Bayesian Approach For 3D Reconstruction From a Single Image
Multimodal Interaction Dr. Mike Spann
Prakash Chockalingam Clemson University Non-Rigid Multi-Modal Object Tracking Using Gaussian Mixture Models Committee Members Dr Stan Birchfield (chair)
EADS DS / SDC LTIS Page 1 7 th CNES/DLR Workshop on Information Extraction and Scene Understanding for Meter Resolution Image – 29/03/07 - Oberpfaffenhofen.
1. Introduction Motion Segmentation The Affine Motion Model Contour Extraction & Shape Estimation Recursive Shape Estimation & Motion Estimation Occlusion.
 In electrical engineering and computer science image processing is any form of signal processing for which the input is an image, such as a photograph.
CSCE 5013 Computer Vision Fall 2011 Prof. John Gauch
Digital Image Processing & Analysis Spring Definitions Image Processing Image Analysis (Image Understanding) Computer Vision Low Level Processes:
Under Supervision of Dr. Kamel A. Arram Eng. Lamiaa Said Wed
AUTOMATIZATION OF COMPUTED TOMOGRAPHY PATHOLOGY DETECTION Semyon Medvedik Elena Kozakevich.
Video Based Palmprint Recognition Chhaya Methani and Anoop M. Namboodiri Center for Visual Information Technology International Institute of Information.
DEVELOPMENT OF ALGORITHM FOR PANORAMA GENERATION, AND IMAGE SEGMENTATION FROM STILLS OF UNDERVEHICLE INSPECTION Balaji Ramadoss December,06,2002.
1 Digital Image Processing Dr. Saad M. Saad Darwish Associate Prof. of computer science.
The University of Texas at Austin Vision-Based Pedestrian Detection for Driving Assistance Marco Perez.
Digital Image Processing & Analysis Fall Outline Sampling and Quantization Image Transforms Discrete Cosine Transforms Image Operations Image Restoration.
Recognizing Action at a Distance Alexei A. Efros, Alexander C. Berg, Greg Mori, Jitendra Malik Computer Science Division, UC Berkeley Presented by Pundik.
Augmented Reality and 3D modelling By Stafford Joemat Supervised by Mr James Connan.
Vision-based human motion analysis: An overview Computer Vision and Image Understanding(2007)
Image Recognition System in Fields National Agriculture and Food Research Organization and University of Tsukuba Kei Tanaka.
1 Research Question  Can a vision-based mobile robot  with limited computation and memory,  and rapidly varying camera positions,  operate autonomously.
COMP322/S2000/L171 Robot Vision System Major Phases in Robot Vision Systems: A. Data (image) acquisition –Illumination, i.e. lighting consideration –Lenses,
Chapter 5 Multi-Cue 3D Model- Based Object Tracking Geoffrey Taylor Lindsay Kleeman Intelligent Robotics Research Centre (IRRC) Department of Electrical.
Jack Pinches INFO410 & INFO350 S INFORMATION SCIENCE Computer Vision I.
Introduction to Related Papers of Vessel Segmentation Methods Advisor : Ku-Yaw Chang Student : Wei-Lu Lin 2015/1/7.
Univ logo Research and Teaching using a Hydraulically-Actuated Nuclear Decommissioning Robot Craig West Supervisors: C. J. Taylor, S. Monk, A. Montazeri.
Colour and Texture. Extract 3-D information Using Vision Extract 3-D information for performing certain tasks such as manipulation, navigation, and recognition.
October 16, 2014Computer Vision Lecture 12: Image Segmentation II 1 Hough Transform The Hough transform is a very general technique for feature detection.
Final Year Project. Project Title Kalman Tracking For Image Processing Applications.
CSSE463: Image Recognition Day 29 This week This week Today: Surveillance and finding motion vectors Today: Surveillance and finding motion vectors Tomorrow:
Robotics Chapter 6 – Machine Vision Dr. Amit Goradia.
Visual Information Processing. Human Perception V.S. Machine Perception  Human perception: pictorial information improvement for human interpretation.
Processing visual information for Computer Vision
Mitsubishi robot arm.
DIGITAL SIGNAL PROCESSING
Digital image self-adaptive acquisition in medical x-ray imaging
Tremor Detection Using Motion Filtering and SVM Bilge Soran, Jenq-Neng Hwang, Linda Shapiro, ICPR, /16/2018.
Digital Image Processing
CMSC 426: Image Processing (Computer Vision)
CSSE463: Image Recognition Day 30
Grape Detection in Vineyards Introduction To Computational and Biological Vision final project Kobi Ruham Eli Izhak.
Introduction to Artificial Intelligence Lecture 22: Computer Vision II
Presentation transcript:

National Research Council Canada Conseil national de recherches Canada National Research Council Canada Conseil national de recherches Canada Institute for Information Technology Visual Information Technology Group Institute for Information Technology Visual Information Technology Group Canada Dr. Dmitry O. Gorodnichy Computational Video Group Institute for Information Technology National Research Council Canada Adding safety to autonomous robot manipulation using video-cameras ( Monitoring of robot motions ) ROSA final workshop. April 17, 2002

2. ROSA: Adding safety using video. (Dmitry Gorodnichy) Setup and Driving Desire -There are many video-cameras on the station -14 cameras on ISS -4 cameras on SSRMS -Can they be used to make operations on the station -easier? -safer? -in particular, the manipulation of SSRMS? -autonomous -manual

3. ROSA: Adding safety using video. (Dmitry Gorodnichy) Cameras and what they can see

4. ROSA: Adding safety using video. (Dmitry Gorodnichy) Application #1 1.Automatically detect the malfunctioning of arm joint encoders, by observing the arm with video- cameras. “Compare what is seen with what should be seen” -Requires -Knowledge and agreement of knowledge of both SSRMS kinematics and camera configuration -Somebody to operate the cameras

5. ROSA: Adding safety using video. (Dmitry Gorodnichy) Application #2 2. Help operator to manipulate the arm, by automatically detecting it in the view of video- cameras “Is there SSRMS in the image? If yes, where it is?” -Requires -Images only -SSRMS description This can be a part of any SSRMS monitoring system

6. ROSA: Adding safety using video. (Dmitry Gorodnichy) Two problems to be resolved Kinematics problem - to find out where the arm should be seen -> FastKin, Cosmos Vision problem - to find out where the arm is actually seen -> R.A.C.E. Major challenge – Recognition Problem Extract the image of SSRMS = Recognize SSRMS in the image

7. ROSA: Adding safety using video. (Dmitry Gorodnichy) Don’t forget: Vision is very ill-posed problem i.e. everything (approaches,complexity,results,success…) depends on the images used

8. ROSA: Adding safety using video. (Dmitry Gorodnichy) R.A.C.E. software -For testing the applicability of different approaches -Uses custom made kinematics & image processing -Integrates: FastKin (MDR) Cosmos (NRC) -Features: joints setup selection camera selection

9. ROSA: Adding safety using video. (Dmitry Gorodnichy) R.A.C.E. allows … Allows one to detect parts of the arm which are possible to see by a camera and -to remove background -to show it by colouring -to obtain the usefulness of the image -to calculate the discrepancy level between what is seen and what should be seen (according to the kinematics) -With different image processing techniques

10. ROSA: Adding safety using video. (Dmitry Gorodnichy) Approaches tested -Intensity discontinuity based (+ gradient filters + Hough transform): assumes edges can be seen and many of them are parallel …………………………………...… good -Texture based segmentation (+ region-growing + shape-from- shading): assumes colour histogram of SSRMS is known ….good -Geometry (Cylinder) based: assumes SSRMS is a collection of cylinders of a known width ……………… very good - Morphology: skeletonization, AND/NOR/NOT of the above approaches …………………………….……. best - Contour-based (snakes): assumes there’s a complete contour ….bad

11. ROSA: Adding safety using video. (Dmitry Gorodnichy)

12. ROSA: Adding safety using video. (Dmitry Gorodnichy) Results and Conclusions Tests conducted on images generated by Cosmos showed that: while it is not always possible to detect completely the arm position, R.A.C.E can significantly facilitate understanding of image context (such as “where SSRMS is”), which can be used 1) for automatic detection of the best views 2) for operator-based manipulation of the arm. Most efficient recognition techniques are determined Possibility of automatic detection of joint malfunctioning is shown (provided that there is the agreement between the kinematics equations and the camera parameters) Ready for experiments with real images.