Adaptive Image Generalized Fuzzy Enhancement and Quality Evaluation
Fa Yi Cui, Lei Lei Ma · Applied Mechanics and Materials · 2014
Before image feature detection and recognition, image enhancement can highlight the main people or things and their details from foreground, and also can suppress the useless information from background effectively. An algorithm model of adaptive image generalized fuzzy enhancement is established. For all aspects of the algorithm model, a variety of computing forms are put forward, and the evaluation standard of image quality is defined. The principle of algorithm is to achieve space transform between image gray space and generalized fuzzy space using generalized membership transform and its adverse transform. In the process of space transform, the contrast among successive region for space of generalized fuzzy membership grade is enhanced by generalized fuzzy enhancement function. Enhanced images are evaluated by quality standard, and the optimal values of adjustable parameters of membership grade transformation function and the fuzzy enhancement function are selected adaptively based on the optimal quality. Then, the enhanced image with best quality can be obtained. Experiments show that the extracted contour of enhanced image is structured, weak-edge-highlighting, and rich-detail.