A method for identifying string-shaped fruit branches based on monocular and binocular cameras

Abstract

The invention provides a method for identifying string-shaped fruit branches based on a monocular camera and a binocular camera. The method first uses a monocular camera to obtain a color image of string-shaped fruit features including fruits, leaves and branches; then uses a support vector machine Train, segment, identify, and extract color images with the AdaBoost algorithm to obtain color images of fruits, leaves, and branches; then use the fruit bunch classification principle to classify fruits into single-fruit bunches, double-fruit bunches, and multi-fruit bunches , and then identify the fruit bunch parent branch through the vertical line of the upper and lower bottom of the fruit bunch circumscribed rectangle; finally, a binocular camera is used to obtain the binocular stereo image of the fruit bunch parent branch, and the normalized cross-correlation function is used to realize the left image and the right The features of the image are matched, and finally the spatial coordinates of the parent branch are obtained. The method can accurately and effectively identify fruit bunches and locate the parent branches of the fruit bunches, realize intelligent picking of bunch-shaped fruits, high picking efficiency, and effectively reduce labor costs.

Description

technical field [0001] The invention relates to the technical field of fruit intelligent picking, in particular to a method for identifying string-shaped fruit branches based on a monocular camera and a binocular camera. Background technique [0002] my country is the main producer of longan, grape, lychee and other bunch-shaped fruits. The total planting and production volume ranks among the top in the world, and has huge economic potential; for example, litchi, as a typical subtropical bunch-shaped fruit, is planted in South China It has become one of the important economic pillars for farmers to get rid of poverty and become rich. [0003] At present, picking string-shaped fruits is basically carried out manually, which is labor-intensive and expensive, and wastes a lot of manpower and material resources. With the advancement of science and technology and the development of intelligence level, there are more and more researches and applications of picking robots, which br...

AI Technical Summary

Problems solved by technology

However, due to the growth characteristics of the bunch-shaped fruit itself, the whole fruit bunch is very easy to grow randomly, which makes it difficult to identify the fruit bunch, the identification and positioning of the mother branch, and the inability to accurately find the entire bunch-shaped fruit. The string-type fruit intelligent picking system faces great challenges in wide application

Images