Tomato string picking point detection method based on RGB-D information fusion

• Main Authors: LI YINUO, WANG JIAN, LUO GAIFANG, HOU YIJUN, SHI YAN, WANG KUN, SUN SHENG, GUO WENHUI, SONG GUOZHU, MA RONGTING, JING CHAO
• Format: Patent
• Language: Chinese; English
• Published: 01.08.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; PHYSICS

The invention discloses a tomato string picking point detection method based on RGB-D information fusion. The tomato string picking point detection method has the advantages of being high in recognition rate, accurate in positioning and the like. After tomato bunch fruit stems are identified through an RGB-D depth camera and a YOLOv7 algorithm, the fruit stems are segmented through a K-means clustering method, a skeletonization algorithm and other algorithms, and the positions of picking points are obtained; and finally, obtaining the depth value of the picking point through a large neighborhood averaging method and conversion correction processing, and completing the obtaining of the image coordinate and depth information of the picking point. The tomato bunch picking device can solve the problems that stems of tomato bunches are blocked, the tomato bunches are too thin or short, and the shapes are irregular, the tomato bunch picking efficiency is improved, and the tomato bunch picking device has great signi