Multi-material positioning method based on image stitching

• Main Authors: LEI SHANSHAN, ZHANG YINTENG, HOU LONGXIAO, WANG TONGFEI, YANG CHENGLIN, LI JIANPU, FAN HUANTONG, MA SHAOQI, ZHAO CONG, LI XIAOPENG, WANG FENG, XIE SHASHA, ZHANG YANJI, BAI BAOKUN
• Format: Patent
• Language: Chinese; English
• Published: 04.06.2024
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention discloses a multi-material positioning method based on image stitching, and relates to the technical field of image processing, and the method comprises the steps: carrying out the feature detection, registration and fusion stitching of a material image, and obtaining a large-view-field image containing a plurality of materials; carrying out contour fitting and centroid calculation on the spliced image to obtain a pixel coordinate of the center of each material; and finally, obtaining actual coordinates of the material through coordinate conversion and camera displacement. According to the invention, accurate positioning of multiple materials in a large view field can be realized, extra erection space is not needed, and the application flexibility and efficiency are better than those of traditional fixed monocular vision. 本发明公开了一种基于图像拼接的多物料定位方法，涉及图像处理技术领域，首先通过对物料图像进行特征检测、配准和融合拼接成包含多个物料的大视场图像；再通过对拼接图像进行轮廓拟合和质心计算得到每块物料中心的像素坐标；最后通过坐标转换和相机位移得出物料的实际坐标。本发明可以实现对大视场多物料的精准定位，无需额外的架设空间，在应用的灵活性和效率上均优于传统的固定