tailieunhanh - Machine Vision - David Vernon Part 10

Tham khảo tài liệu 'machine vision - david vernon part 10', kỹ thuật - công nghệ, cơ khí - chế tạo máy phục vụ nhu cầu học tập, nghiên cứu và làm việc hiệu quả | Introduction to image understanding matching or feature extraction and classification paradigms are wholly inadequate. To be able to proceed from raw two-dimensional images of the world to explicit three-dimensional structural representations we must adopt a much more sophisticated stance and we must acknowledge that no single process or representation is going to be generally adequate. Thus there is one central theme which runs through the current and now conventional approach to image understanding that we require intermediate representations to bridge the gap between raw images and the abstracted structural model. These representations should make different kinds of knowledge explicit and should expose various kinds of constraint upon subsequent interpretations of the scene. It is the progressive integration of these representations and their mutual constraint to facilitate an unambiguous interpretation of the scene that most characterizes the branch of vision known as image understanding. It is perhaps useful to note that most of the progress that has been made in the past few years has not in fact been in this area of representation integration or data fusion as it is commonly known but rather in the development of formal and well-founded computational models for the generation of these representations in the first place. As we shall see this is no accident. Nevertheless there remains a great deal of work to be done in the area of data fusion. In summary then we can characterize image understanding as a sequence of processes concerned with successively extracting visual information from one representation beginning with digital images organizing it and making it explicit in the representation to be used by other processes. From this perspective vision is computationally modular and sequential. What we must now proceed to look at are the possible organizations of visual processes the representations and the visual processes themselves. We will begin with the .