Recognition of Garlic Cloves Orientation Based on Machine Vision

摘要

In order to realize the garlic single grain precision seeding, keep the garlic in an upright position, ensure the state of the garlic cloves upward, this paper uses machine vision technology, the collected garlic image preprocessing, image segmentation, expansion, area recognition method to process garlic images. Using the Otsu algorithm to segment the image, we can separate the garlic image from the background very well. Based on the asymmetry of the area between the upper part and the lower part of the garlic. The direction of the garlic is identified by calculating the area of two parts. The experimental results show that the recognized accuracy of garlic's direction could reach to 95.33% by using this algorithm. And it can provide technical support for future single-grain precision planting of garlic.