Height information acquisition method of seedling with machine vision

In plant factory, transplanting as an important step of nursery seedling process, it is necessary to achieve its automation and intelligence. To improve the survival rate of seedling transplanting, the fitness of transplanting seeding need to be distinguished. Seedling height, an important indicator of transplanting fitness, this paper attempts to use machine vision identify technology to analysis and judgment the height information rapidly to meet the transplanting requirements. In this paper, with the color images of pepper seedling as sample, the main stem characteristics of seedling were extracted out by using image processing algorithms. Then, the key points of every potted-seedling trunk were extracted by a Harris corner detection algorithm. The fitting line was obtained by the weighted least-squares linear fitting with the key points, and found out the maximum y-coordinate difference of all corners coordinates in each strain of potted-seedling. The average relative deviation algorithm of Harris corner detection algorithm with principal axis method was 2.85%