双目立体视觉技术在果蔬采摘机器人中的应用

Binocular stereo vision applied to harvesting robots

采用双目立体视觉技术获取温室内果蔬的三维位置信息,用于指导果蔬采摘机器人进行自动化采摘作业.试验以植株上成熟的番茄作为研究对象,利用立体摄像机获取不同距离的成熟番茄的立体图像对,通过对图像进行灰度图像处理,将彩色图像转换为灰度图像.然后根据灰度图像对中像素点的相关性进行立体匹配,计算像素点的位置信息而获得一幅深度图像.最后对照番茄形心在深度图像中的位置,获取番茄的三维位置信息.试验结果表明,立体摄像机与番茄的距离小于1 000 mm时,番茄的距离误差为±20 mm.

A measurement of 3 D locations of ripe tomatoes by binocular stereo vision was done for harvesting tomatoes in greenhouse. In this method, a pair of stereo images were obtained by stereo cameras, which were then transformed to gray images. According to the gray correlation, corresponding points of stereo images were searched, and a depth image was obtained by calculating distances between tomatoes and stereo cameras based on triangulation principle. The center of tomato was extracted by distinguishing a tomato from background with image processing. The 3D locations of ripe tomatoes were obtained by comparing coordinate values of the center of tomato ±20 mm when distance is less than 1 000 mm. with the depth image. The error of depth is within