| ECVision indexed and annotated bibliography of cognitive computer vision publications This bibliography was created by Hilary Buxton and Benoit Gaillard, University of Sussex, as part of ECVision Specific Action 8-1 The complete text version of this BibTeX file is available here: ECVision_bibliography.bib
V. Murino and G. Vernazza
ABSTRACT
The growing computational power of current computer systems and the reduced costs of image acquisition devices allow a very large audience to have an easier access to image analysis issues. Artificial neural networks (ANNs), with their massive parallelism, have shown considerable promise in a wide variety of application areas, not related to vision problems only, and have been particularly useful in solving problems for which traditional techniques have failed or proved inefficient. Accordingly, even hard image-analysis applications, like some robotic operations, visual inspection, remote sensing, autonomous vehicle driving, automated surveillance, and many others, have been approached using neural networks (NNs). The investigation of ANNs in this context has received a great deal of attention, on the one hand, as imaging tasks are computationally intensive and high performances potentially can be reached in real time, and on the other hand, thanks to the versatility of neural approaches. Versatility means that a set of interesting properties are shared by neural systems: apart from the inherent parallelism, they allow a distributed representation of the information, easy learning by examples, high generalization ability, and a certain fault tolerance. All these features are inherited by the biological NNs from which they take inspiration. One of the other reasons for the success of ANNs is the plethora of possible architectures that allow almost any imaging problem to be tackled and solved. There is a vast literature on NNs: it deals with both theoretical investigations of their inherent mechanisms, and solutions to real problems. Typically, NNs have been shown to be particularly suitable (and have been used extensively) for pattern recognition problems, namely classification, clustering, and feature selection. Handwritten character recognition is one of the most classical applications in this area, but ANNs are also utilized for many other tasks, like optimization, prediction, control, and, more recently, data mining and information retrieval. The success of NNs is also confirmed by the commercial products currently available, as well as by a number of patents. The purpose of this special issue is to put another brick on this high wall, as it collects papers presenting original research work on applications of ANNs to image analysis and computer vision. It includes ten papers, dealing with recognition and pose estimation, texture analysis, segmentation, image compression, color representation and approximation, and classification. |
|
| Site generated on Friday, 06 January 2006
|