Research on image processing based on pattern recognition

Abstract

Image processing is to transform the image into digital matrix by mathematical formula, then store the data in the computer, and enhance the source bitmap by computer, including gray-scale processing, median filtering and binarization. Then the image is cut, including contour edge processing, seed filling and so on. In this paper, the image is recognized and processed based on pattern recognition, and then a pre classifier is designed to build a system model of fruit intelligent classification. After the fruit is taken by industrial camera, the corresponding image processing and feature value extraction are carried out. Then the extracted eigenvalues are used for classification. Through this method to achieve a variety of fruit classification effect.