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


Machine Intelligence Unit (MIU-Kolkata)


Pal, Sankar Kumar (MIU-Kolkata; ISI)

Abstract (Summary of the Work)

During the last five decades or even more a large number of researchers are trying to design intelligent systems to perform tasks at which human beings are more efficient at present. One of the most important behavioral tasks in which human beings show their expertise is image analysis or recognition; where a large amount of pictorial data is processed in a very small amount of time (called real time). Widespread attempts have been made to develop intelligent systems (under different names, like pattern recognition system, image under- standing system, computer vision system etc.) for pictorial pattern analysis and recognition. The importance of these subjects is due to their extensive use in a broad range of scientific and commercial application areas such as :• document processing : recognition of printed and hand written characters;• biology and medicine : cytology (blood cell counting, chromosome analysis, etc.), radiology, ECG analysis, EEG analysis and cancerous growth detection;• criminal identification : finger print analysis, hand written character recognition and face recognition;• military applications : target detection, missile guidance and airport detection;• industrial automation : robotics, micro graphic image analysis and non- destructive testing;• remote sensing : land surface determination, carth surface classification, radio astronomy, meteorology and environment monitoring;• man-machine communication : processing of visual information for image understanding and scene analysis.In addition to the classical mathematical techniques (deterministic and prob- abilistic) new technologies, like fuzzy set theory and artificial neural network models are suggested for designing such systems. The theory of fuzzy sets pro- vides an approximate and yet effective means for describing the characteristics of a system which is too complex or ill-defined to admit of precise mathematical analysis. Fuzzy sets can model the human thinking process and behavior, and is reputed to handle (to a reasonable extent) uncertainties arising from deficiencies of information available from a situation, the deficiencies may result from incomplete, ill-defined, not fully reliable, vague and contradictory information.Artificial neural networks are signal processing systems that try to emulate the human brain, i.e., the architecture and working principle of human nervous systems, by providing a mathematical model of combination of numerous neurons connected in a network. These models are reputed to have the following characteristics : adaptivity, speed, fault tolerance and optimality.The aim of this thesis is to present some results of investigations, both theoretical and experimental, that demonstrate the effectiveness of some of the concepts of fuzzy set theory and artificial neural network models in formulating various image segmentation/object extraction methodologies for image analysis. The rest of this chapter is organized as follows. In section 1.2 preliminaries of image processing and analysis are described in short. A survey on existing image segmentation techniques is given in section 1.3. A discussion on the integration of the merits of fuzzy set theory and neural networks is made in section 1.4. Scope of the thesis is provided in section 1.5.


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