Orientation selection is the inference of orientation information out of images. It is one of the foundations on which other visual structures are built, since it must precede the formation of contours out of pointill...
详细信息
Orientation selection is the inference of orientation information out of images. It is one of the foundations on which other visual structures are built, since it must precede the formation of contours out of pointillist data and surfaces out of surface markings. We take a differential geometric view in defining orientation selection, and develop algorithms for actually doing it. The goal of these algorithms is formulated in mathematical terms as the inference of a vector field of tangents (to the contours), and the algorithms are studied in both abstract and computational forms. They are formulated as matching problems, and algorithms for solving them are reduced to biologically plausible terms. We show that two different matching problems are necessary, the first for 1-dimensional contours (which we refer to as Type I processes) and second for 2-dimensional flows (or Type II processes). We conjecture that this difference is reflected in the response properties of “simple” and “complex” cells, respectively, and predict several other psychophysical phenomena.
During the last few years motion analysis has developed into a major field of interest in image analysis and understanding. Considerable progress has been made as evidenced by the number of publications. However, most...
详细信息
During the last few years motion analysis has developed into a major field of interest in image analysis and understanding. Considerable progress has been made as evidenced by the number of publications. However, most of these have been confined to the study of rigid body motion. Very few have addressed the issue of non-rigid body motion which poses many interesting and difficult problems. We propose a new technique for analyzing non-rigid body motion of closed boundary shapes that is based on the computation of skeletons at multiple resolutions. This technique is used for analyzing the structural changes in the morphology of locomoting lymphocytes and, in particular, of their pseudopods.
This paper presents a multi-robot system which assembles integrated circuits. The problems addressed are the calibration of the robotic station, the control of several robots working concurrently, and pattern recognit...
详细信息
This paper presents a multi-robot system which assembles integrated circuits. The problems addressed are the calibration of the robotic station, the control of several robots working concurrently, and pattern recognition. These problems are discussed in the context of a distributed system. The use of image processing is omnipresent. For instance, vision feedback is used to acknowledge almost every move of the robots and also to calibrate the camera itself. Pattern recognition techniques are employed to detect the exact orientation of an IC die. Several robots are used in order to give more flexibility to the system and allow us to acquire experience which could lead toward a multirobot multitasking system.
The authors present a new method for locating discontinuities in one-dimensional cuts through idealized images. The model for such a cut is a sampled, piecewise smooth (C**1) function corrupted by noise. The new metho...
详细信息
ISBN:
(纸本)0818605456
The authors present a new method for locating discontinuities in one-dimensional cuts through idealized images. The model for such a cut is a sampled, piecewise smooth (C**1) function corrupted by noise. The new method uses an extension of the mathematical definition that a function is discontinuous at a point if the left- and right-hand limits of the function (or any of its derivatives) are unequal. The extended definition is that a noisy function is discontinuous at a point if the difference in the estimates of the limits is statistically significant. It follows from this extended definition that locating discontinuities must be carried out at the same time as determining the structure of the uncorrupted function in a local neighbourhood about the discontinuity (we call this the local structure of the discontinuity). Determining this local structure is an important first step in interpreting the physical event causing the image discontinuity.
A scheme is presented by which a precise visual image can be reconstructed from a combination of these multiple-size representations. The scheme, which supports the position that image analysis begins in the cortex, d...
详细信息
ISBN:
(纸本)0818605456
A scheme is presented by which a precise visual image can be reconstructed from a combination of these multiple-size representations. The scheme, which supports the position that image analysis begins in the cortex, differs in basic technical ways from other schemes proposed to account for the communication gap and hyperacuity. One such class of schemes is based on (sin chi )/ chi reconstruction filters. The numerical analysis of such filters shows that they require too much spatial support (i. e. , several lobes on either side) to function properly with the limited support apparently available. The present scheme is based on local (Hermite) polynomials. A completely different theoretical position is that the receptive field operators are participating in 'edge detection', but both computational experiments and mathematical analysis provide evidence that this position in untenable.
作者：
Kamal KantZucker, StevenMcGill Univ
Computer Vision & Robotics Lab Montreal Que Can McGill Univ Computer Vision & Robotics Lab Montreal Que Can
In this paper, the authors present a novel approach to solving the trajectory planning problem (TPP) in time-varying environments. The essence of our approach lies in a decomposition of the TPP into two sub-problems: ...
详细信息
ISBN:
(纸本)0892525568
In this paper, the authors present a novel approach to solving the trajectory planning problem (TPP) in time-varying environments. The essence of our approach lies in a decomposition of the TPP into two sub-problems: (i) planning the path to avoid collision with the static obstacles, and then (ii) planning the velocity along the path to avoid collision with the moving obstacles. The authors call the first sub-problem, the path planning problem (PPP), and the second, the velocity planning problem (VPP). Thus, the decomposition is summarized by the equation TPP equals PPP plus VPP. A variation of an existing approach to solve the PPP is used. They transform the VPP to a 2-dimensional PPP in path-time space. The resulting 2-dimensional PPP is then solved to determine the collision-free velocity profile.
In this paper, the authors generalize the path planning problem to that of the trajectory planning problem (TPP) in a time-varying environment. They present an algorithm to solve a special case of the TPP - the fixed ...
详细信息
ISBN:
(纸本)0818605456
In this paper, the authors generalize the path planning problem to that of the trajectory planning problem (TPP) in a time-varying environment. They present an algorithm to solve a special case of the TPP - the fixed path TPP. In this case, the path is a point set fixed in space and the aim is to determine the velocity of a robot vehicle moving along this path from an initial position to a final position while avoiding moving obstacles. The solution is essentially equivalent to a (static) 2-D path planning problem. Graph searching techniques are used to construct the velocity profile.
This volume of Springer’s Lecture Notes in computer Science series comprises th the scienti?c proceedings of the 10 International Workshop on Digital Mammography (IWDM), which was held June 16–18, 2010 in Girona, Ca...
详细信息
ISBN:
(数字)9783642136665
ISBN:
(纸本)9783642136658
This volume of Springer’s Lecture Notes in computer Science series comprises th the scienti?c proceedings of the 10 International Workshop on Digital Mammography (IWDM), which was held June 16–18, 2010 in Girona, Cata- nia. The IWDMmeetingstraditionallybringtogetheradiversesetofresearchers (physicists, mathematicians, computer scientists, engineers), clinicians (radi- ogists, surgeons) and representatives of industry, who are jointly committed to developing technology, not just for its own sake, but to support clinicians in the early detection and subsequent patient management of breast cancer. The IWDM conference series was initiated at a 1993 meeting of the SPIE Medical Imaging Symposium in San Jose, CA, with subsequent meetings hosted every two yearsbyresearchersaroundthe world. Former workshopswereheld in York, England (1994), Chicago, IL USA (1996), Nijmegen, The Netherlands (1998), Toronto, Canada (2000), Bremen, Germany (2002), Durham, NC, USA (2004), Manchester, UK (2006) and Tucson, AZ USA (2008). Each of these scienti?c events was combined with very successful and focused industrial and research exhibits, which demonstrated the milestones of digital mammography over the years. A total number of 141 paper submissions from 21 countries were received. Each of these four-page abstract submissions was reviewed in a blind process by at least two members of the Scienti?c Committee, which led to a ?nal selection of 46 oral presentations and 57 posters during the two and one-half days of scienti?c sessions.
The ability of a robot to build a persistent, accurate, and actionable model of its surroundings through sensor data in a timely manner is crucial for autonomous operation. While representing the world as a point clou...
详细信息
The ability of a robot to build a persistent, accurate, and actionable model of its surroundings through sensor data in a timely manner is crucial for autonomous operation. While representing the world as a point cloud might be sufficient for localization, denser scene representations are required for obstacle avoidance. On the other hand, higher-level semantic information is often crucial for breaking down the necessary steps to autonomously complete a complex task, such as cooking. So the looming question is, What is a suitable scene representation for the robotic task at hand? This survey provides a comprehensive review of key approaches and frameworks driving progress in the field of robotic spatial perception, with a particular focus on the historical evolution and current trends in representation. By categorizing scene modeling techniques into three main types—metric, metric–semantic, and metric–semantic–topological—we discuss how spatial perception frameworks are transitioning from building purely geometric models of the world to more advanced data structures incorporating higher-level concepts, such as the notion of object instances and places. Special emphasis is placed on approaches for real-time simultaneous localization and mapping, their integration with deep learning for enhanced robustness and scene understanding, and their ability to handle scene dynamicity as some of the hottest topics of interest driving robotics research today. We conclude with a discussion of ongoing challenges and future research directions in the quest to develop robust and scalable spatial perception systems suitable for long-term autonomy.
暂无评论