Integrating Metaheuristics in Computer Vision for Real-World Optimization Problems

17Handwritten Image Enhancement Based on Neutroscopic-Fuzzy and K-Mean Clustering

Jaspreet Kaur¹, Divya Gupta¹*, Simarjeet Kaur² and Amrinder Singh³

¹Department of Computer Science and Engineering, Chandigarh University, Mohali, India

²Chitkara University Institute of Engineering and Technology, Chitkara University, Punjab, India

³Department of History, Chandigarh University, Mohali, India

Abstract

Image processing is the process of applying different operations to an image in order to either improve it or extract useful information from it. In this chapter, we are addressing the problems that are associated with handwritten document with the passage of time. Hand-written documents can deteriorate for a number of reasons, including the original document’s quality, the ink or paper used, and the document’s age. As a result, the text could become unclear, blurry, or faded, making it challenging to read and comprehend. Identification and recognition of old manuscripts is very difficult from low resolution images due to poor boundary and contrast. Work done in this field deals with recognizing content of old manuscripts. Many documents that are produced today, such as personal letters, pictures, agreements, newspapers and medical records serve as valuable historical documents for some people in the future. Further, most of them will be vanished in the future because many documents are often made from faulty materials that disappear, tear or destroy over time or are even stored ...