A machine vision system for seeds quality evaluation using fuzzy logic

This paper presents an automatic system for monitoring seed germination by means of a software tool incorporating an artificial vision system and a fuzzy logic-based classifier. The system employs a colour CCD camera to image trays in which seeds have been planted in cells and an image processing system to identify the seedlings and their leaf area. This information enters a fuzzy logic-based classifier that simulates the quality grading of seedlings undertaken by experienced technicians, based on their growth pattern. In some tests performed with three horticultural crops (lettuce, cauliflower and tomatoes), the fuzzy logic closely matched the evaluations made by technicians, thereby showing its potential to relieve them of a time-consuming task. The system's associated database provides many possibilities for decision support to the seed production industry.