Soyabean Seed Boundary Detection Algorithm

Abstract

Glycine max (soyabean) commonly used for oil, soya products, fodder, and protein supplements are inseparable part of human life and therefore its evaluation for high quality to ensure consistent increase had gain importance. In this work, extracting the boundary of soyabean seed from the background is proposed. The images are taken on a white background and using image processing tools and morphological operations the boundary of the seed region is automatically marked and then cropped for analysis. The proposed soybean quality inspection system can be used as an effective tool for real-time online inspection of soybean quality. Image processing and machine learning technique are modified to use the quality grading of soybean seeds. Due to quality grading is a very important process for the soybean industry and soybean farmers. There are still some critical problems that need to be overcome. Therefore, the key contributions of this paper are first, a method to eliminate shadow noise for segment soybean seeds of high quality. Keywords: Soybean Quality, Crushing Rate, Impurity Rate, Image Processing, Improved U-Net Algorithm.