The authors present the organization of an interleaved wrap-around memory system for a partitionable parallel/pipeline architecture with P pipes of L processors each. The architecture is designed to efficiently suppor...
详细信息
ISBN:
(纸本)0818690895
The authors present the organization of an interleaved wrap-around memory system for a partitionable parallel/pipeline architecture with P pipes of L processors each. The architecture is designed to efficiently support real-time imageprocessing and computer vision algorithms, especially those requiring global data operations. The interleaved memory system makes the architecture highly scalable in that L and P can be chosen to optimize performance for particular problems and reconfigurable in that, once L and P are fixed, problems of any size can still be mapped onto the architecture. The authors demonstrate techniques and methods for mapping computational structures to the architecture by considering the case of the 1-D butterfly network (1DBN). Since many other computational structures can be mapped to 1DBN, this gives a firm application base for the architecture. The authors also demonstrate methods for scheduling and controlling the memory system.
The problem of automatically learning knowledge-directed control strategies is considered. In particular, this work addresses the problem of learning object-specific recognition strategies from object descriptions and...
详细信息
ISBN:
(纸本)0818620579
The problem of automatically learning knowledge-directed control strategies is considered. In particular, this work addresses the problem of learning object-specific recognition strategies from object descriptions and sets of interpreted training images. A separate recognition strategy is developed for every object in the domain. The goal of each recognition strategy is to identify any and all instances of the object in an image, and give the 3-D position (relative to the camera) of each instance. The goal of the learning process is to build a strategy that minimizes the expected cost of recognition, subject to accuracy constraints imposed by the user.
Template matching is concerned with measuring the similarity between patterns of two objects. A massively parallel approach to patternrecognition with a large template set is proposed. A class of imagerecognition pr...
详细信息
ISBN:
(纸本)0818620307
Template matching is concerned with measuring the similarity between patterns of two objects. A massively parallel approach to patternrecognition with a large template set is proposed. A class of imagerecognition problems inherently needs large template sets, for example, the recognition of Chinese characters, which needs thousands of templates. The proposed algorithm is based on the SIMD-SM-R machine or the SIMD machine with broadcasting abilities, which is the most popular parallel machine to date, and uses a multiresolution method to search for the matching template. The approach uses the pyramid data structure for the multiresolution representation of templates and the input imagepattern. For a given image it scans the template pyramid searching for the match. Implementation of the proposed scheme is described.
An image-matching algorithm, called the dynamic space-warping algorithm (DSWA), is presented. It is based on both local-distance diagrams and dynamic programming. The DSWA can solve space-warping problems (e.g., shrin...
详细信息
An image-matching algorithm, called the dynamic space-warping algorithm (DSWA), is presented. It is based on both local-distance diagrams and dynamic programming. The DSWA can solve space-warping problems (e.g., shrinking, enlarging, rotation, and distortion) with good performance by embedding controllable flexibility (or warping). The concept of flexibility can be explained using local-distance diagrams. With flexibility, the local-distance diagram between two two-dimensional images is four dimensional. Based on compression and expansion, DSWA generates a minimum distance from the four-dimensional local-distance diagram. Experimental results show that the DSWA is very reliable.
A two-step algorithm is presented for thinning noisy images. The first step consists of removing noisy spurs from an image using a combination of morphological operations. In the second step, thinning is performed by ...
详细信息
A two-step algorithm is presented for thinning noisy images. The first step consists of removing noisy spurs from an image using a combination of morphological operations. In the second step, thinning is performed by using the standard morphological skeleton transform (MST). This approach is useful in model-based vision applications where the features of the image are required for object recognition. The use of this algorithm in sensor array imaging situations is demonstrated.
An adaptive-control strategy developed for multidiameter rollers is described. The method is based on microcomputer-controlled in-process size-measurement techniques. The system enables faster and more accurate roller...
详细信息
ISBN:
(纸本)081861966X
An adaptive-control strategy developed for multidiameter rollers is described. The method is based on microcomputer-controlled in-process size-measurement techniques. The system enables faster and more accurate roller grinding, replacing the skilled operator with an expert system. Stepper-motor-driven steadies and the associated computer control system are described.
A network is described that recognizes objects from uncertain image-derivable descriptions. The network handles uncertainty by making the recognition and segmentation decisions simultaneously, in a cooperative way. Bo...
详细信息
ISBN:
(纸本)0818620579
A network is described that recognizes objects from uncertain image-derivable descriptions. The network handles uncertainty by making the recognition and segmentation decisions simultaneously, in a cooperative way. Both problems are posed as labeling problems, and a coupled Markov random field (MRF) is used to provide a single formal framework for both. Prior domain knowledge is represented as weights within the MRF network and interacts with the evidence to yield a labeling decision. The domain problem is the recognition of structured objects composed of simple junction and link primitives. Implementation experiments demonstrate the parallel segmentation and recognition of multiple objects in noisy ambiguous scenes with occlusion.
A landmark recognition system is proposed for navigation applications. This system uses artificial neural nets to classify patterns in multiple feature spaces characterized by different sensors. The core classifier, a...
详细信息
A landmark recognition system is proposed for navigation applications. This system uses artificial neural nets to classify patterns in multiple feature spaces characterized by different sensors. The core classifier, a bidirectional associative memory, has been partially simulated by a computer program and initial results show satisfactory performance.
A boundary descriptor called surface code is developed. It can be used as a matching feature in the process of identification of an object. The surface code generation technique was tested on both real and synthetic i...
详细信息
A boundary descriptor called surface code is developed. It can be used as a matching feature in the process of identification of an object. The surface code generation technique was tested on both real and synthetic images. The surface codes have been generated successfully for all the images studied. The results indicate that the code is orientation independent and dimension independent, in which case hierarchical information in the form of extra code elements is needed to distinguish between similar shapes of different sizes.
An examination is made of the problem of selecting models from a library, and the combinatorics of determining that a candidate object is not present in the data are examined. It is shown that the expected search is a...
详细信息
ISBN:
(纸本)0818620579
An examination is made of the problem of selecting models from a library, and the combinatorics of determining that a candidate object is not present in the data are examined. It is shown that the expected search is again exponential, implying that naive approaches to indexing are likely to carry an expensive overhead, since exponential work is needed to weed out each incorrect model. The analytical results are shown to be in agreement with empirical data for cluttered object recognition.
暂无评论