Affiliation:
1. Uttarakhand Technical University, India
2. Jaypee University of Engineering and Technology, India
3. Indian Institute of Forest Management, India
Abstract
An image may be mist full for the one or may be nostalgic for the other. But for a researcher, each image is distinguished on the basis of its low-level features like color, shape, size. Other features like edges, corner/interesting points, blobs/region of interest, ridges, etc. can also be used for computation purpose. Using these features distinctions, an image can be processed for the purpose of enhancing the images having same features, matching, and shortlisting of similar images from a random available image database. This could be done using soft computing techniques like neural networks, fuzzy logic, and evolutionary computation methods. Neural networks can participate effectively in image processing in several ways.