The machine vision community has expended tremendous effort in the research and development of algorithms in an effort to develop a system that is capable of seeing the world as humans do. These algorithms often focus...
详细信息
The machine vision community has expended tremendous effort in the research and development of algorithms in an effort to develop a system that is capable of seeing the world as humans do. These algorithms often focus on the accomplishment of specific tasks analogous to human vision such as scene awareness, object detection, object recognition, and object tracking. Joining forces with cognitive neuroscientists has steered much of this research towards the development of algorithms that not only accomplish the required tasks, but endeavor to do so in a biologically inspired fashion. Still, development and evaluation of these so-called neuromorphic algorithms is often done in isolation, with little regard given to the rest of the system necessary to make this human-like system a reality. This dissertation provides a framework for the current and future development of complex and highly integrated multi-algorithm vision systems. This framework not only enables the composition of such systems, but enables seamless development and integration of improved algorithmic modules. In addition to this high-level system composition framework, the Cerebrum tool, targeted at development of hardware-accelerated architectures is detailed in this work. This tool enables the creation of such hardware-based accelerators by researchers and engineers without specific or detailed knowledge of the target hardware *** addition to the framework and tools, this dissertation also details the analysis, development and evaluation of hardware accelerators for HMAX object recognition and AIM saliency detection. Armed with this intelligent framework and algorithmic accelerators, demonstrations of vision systems that leverage multiple algorithms are constructed and *** object classification, leveraging the benefits of Exemplar SVM and accelerated HMAX is shown to provide performance superior to either algorithm in isolation. Furthermore, a more complex system, targeting th
Gestures are used in day to day life like nodding and waving without us being aware of them. It has become an important part in the communication among the humans. In the recent years new methods of Human Computer Int...
详细信息
ISBN:
(纸本)9781467357586;9781467357593
Gestures are used in day to day life like nodding and waving without us being aware of them. It has become an important part in the communication among the humans. In the recent years new methods of Human Computer Interaction (HCI) are being developed. Some of them are based on interaction with machines through hand, head, facial expressions, voice, touch and many are still the current topic of research. However relying on just one of them reduces the accuracy of the whole HCI and is also limiting the options available to users. The objective of this paper is thus to use two of the important modes of interaction hand and head to control any application running on computer using Computer vision algorithms. From input video stream, hand is segmented and the corresponding gesture is being recognized based on the shape and pattern of movement of hand. For head gesture recognition, head is first detected and then optical flow method is used to get the movement of head which is then recognized by finite state automata. Using the user interface of the software, an operator can control any interactive application (say VLC player, Image browser etc) using hand and head gestures which in turn are automatically mapped to the mouse and keyboard events through Windows API. The proposed multimodal approach is particularly useful to communicate with computers and other electronic appliances from a distance where mouse and keyboard are not convenient to work with.
Gestures are used in day to day life like nodding and waving without us being aware of them. It has become an important part in the communication among the humans. In the recent years new methods of Human Computer Int...
详细信息
ISBN:
(纸本)9781467357593
Gestures are used in day to day life like nodding and waving without us being aware of them. It has become an important part in the communication among the humans. In the recent years new methods of Human Computer Interaction (HCI) are being developed. Some of them are based on interaction with machines through hand, head, facial expressions, voice, touch and many are still the current topic of research. However relying on just one of them reduces the accuracy of the whole HCI and is also limiting the options available to users. The objective of this paper is thus to use two of the important modes of interaction - hand and head to control any application running on computer using Computer vision algorithms. From input video stream, hand is segmented and the corresponding gesture is being recognized based on the shape and pattern of movement of hand. For head gesture recognition, head is first detected and then optical flow method is used to get the movement of head which is then recognized by finite state automata. Using the user interface of the software, an operator can control any interactive application (say VLC player, Image browser etc) using hand and head gestures which in turn are automatically mapped to the mouse and keyboard events through Windows API. The proposed multimodal approach is particularly useful to communicate with computers and other electronic appliances from a distance where mouse and keyboard are not convenient to work with.
This paper reviews developments since 2000 in the application of machine vision to food and agriculture. The subject involves applying radiation of various wavelengths to materials in order to find more about them: of...
详细信息
This paper reviews developments since 2000 in the application of machine vision to food and agriculture. The subject involves applying radiation of various wavelengths to materials in order to find more about them: often this means looking not only at surfaces but also at internal structures. While visible light frequently provides enough useful information to make sound judgements, with the advent of dual energy X-ray (DEXA) detection, X-rays have been increasingly valuable. Perhaps the most exciting development is the 'spectral image cube' as an investigative tool. There have also been valuable developments in the use of three-dimensional methods, such as 'double Hough transforms' for the accurate delineation of crop rows, so that 'precision agriculture' can be realized, and the use of sets of visual calibration points so that robot vehicles can determine their exact locations and headings. Overall, the steady development of useful vision algorithms has been well matched by the capability of today's computers to implement them at sufficiently high speeds to make them viable.
A methodology is proposed for the computation of output-data errors from given input-data errors for a given computational precision for vision algorithms. The approach is based on replacement of the arithmetic types ...
详细信息
A methodology is proposed for the computation of output-data errors from given input-data errors for a given computational precision for vision algorithms. The approach is based on replacement of the arithmetic types of the object-oriented language used for coding of the tested algorithm by structures describing the values and error models of the numbers represented, The method is useful both for the initial testing of any vision algorithm and for the definitive testing of algorithms for which explicit error models are not known. It is also shown how two error models, namely that of the min/max error propagation and that of local-function linearisation, can be used for assessing the errors introduced by the input data and by the hardware architecture adopted.
The notion of polygonal entropy is introduced which captures the degree of nonconvexity of a polygon through a measure of polygonal visibility. We show that the entropy of a simple polygon is maximal if and only if it...
详细信息
The notion of polygonal entropy is introduced which captures the degree of nonconvexity of a polygon through a measure of polygonal visibility. We show that the entropy of a simple polygon is maximal if and only if it is convex. When ‘normalized’ the entropy measure is a number between ε (a small positive nonzero number) and 1 providing a convenient index which represents the degree of irregularity (nonconvexity) of a simple polygon. A similar but less computationally burdensome measure is also proffered. Suggested uses of the measure are: the evaluation of computational complexity of vision algorithms, pattern recognition and architectural space planning.
暂无评论