Date of Submission


Date of Award


Institute Name (Publisher)

Indian Statistical Institute

Document Type

Doctoral Thesis

Degree Name

Doctor of Philosophy

Subject Name

Computer Science


Computer Vision and Pattern Recognition Unit (CVPR-Kolkata)


Chaudhuri, Bidyut Baran (CVPR-Kolkata; ISI)

Abstract (Summary of the Work)

The important visual characteristics of an object are shape, size, color, brightness, contrast and texture. Of them, shape is a multidimensional concept that is difficult to define. It takes different meanings in different contexts. We try to explain it in terms of their attributes like elongation, roundness, and symmetry: although these terms do not capture the complete notion of shape.Perhaps Gestalt theory Koffka 351 is the first attempt to study the principles of visual perception in a systematic manner. The central concept of this theory is Gestalt' which means form or configuration. In this theory form is examined from physical. psychological and behavioral points of view. Some basic Gestalt principles are discussed in the next section. These principles help us to understand the notion of shape.The introduction of information theory provided another tool for the study of shape. Attneave and others showed how and where the information in the shape of a particular object is concentrated (Attneave 591. Gradually, quantitative approaches of evaluation of shape property were evolved with the advent of digital computers. Controlled synthesis of shape and forms were also considered side by side for graphical applications. An overview of shape analysis is presented in Section 1.1.Our thesis concerns the shape of dot patterns. The dot pattern shape is introduced in Section 1.2 where different approaches of shape computation are surveyed. The shape detection of dot pattern may also be viewed as a set estimation problem. In that case the consistency criterion become an important issue. In this thesis we have considered the set estimator aspect as well. Moreover, the clustering problem is dealt with for the shape extraction from complex point pattern.The scope and layout of the thesis are presented in Section 1.3.


ProQuest Collection ID:

Control Number


Creative Commons License

Creative Commons Attribution 4.0 International License
This work is licensed under a Creative Commons Attribution 4.0 International License.


Included in

Mathematics Commons