A frequently expressed view in image analysis is that a general purpose vision system must be capable of generating rich descriptions without knowledge of specific objects in a scene. Defining the capabilities of such systems has received considerable research attention in the past several years. It is recognized that issues of system engineering play an essential role in the successful development of large computer vision systems. Moreover, it has been observed that computational issues such as architectural limitations, file structures, etc., have to some extent impeded progress in other areas of computer vision [Redd78]. Pragmatic aspects of computer vision have been extensively discussed in connection to basic image analysis capabilities such as edge detection, shape representation, segmentation, etc. [Erma78, Etch83, Neva78]
KeywordsManifold Radar Hull Autocorrelation Convolution
Unable to display preview. Download preview PDF.