Night identification method of apple harvesting robot

Abstract

The invention provides a night identification method of an apple harvesting robot. The method includes the following steps that: image acquisition is performed under an artificial light source condition; color-based image segmentation is performed through adopting a R-G color difference segmentation method, and binaryzation is performed through adopting an OTSU algorithm; noises in images obtained in the second step are filtered out through filtering out small combination regions; approximate positions of apples in the images are positioned, and regions in original images, where the apples are located, are extracted; highlight light-reflection regions of the surfaces of the apples are extracted; and finally, the high highlight light-reflection regions and the images obtained in the third step are added together and are fused, so that final results can be obtained. With the night identification method of the apple harvesting robot adopted, problems existing in light source selection and layout as well as problems caused by shadows and light reflection in the night identification of the apple harvesting robot can be solved, and the apple harvesting robot can work normally at night.

Description

technical field [0001] The invention relates to the field of image recognition of fruit tree picking robots, in particular to a nighttime recognition method for apple picking robots. Background technique [0002] my country is a large apple producing country, and the labor force used for fruit picking accounts for more than 1 / 3 of the labor force used in the entire apple production process. Using apple picking robots for picking can not only reduce labor costs, but also reduce the damage rate and improve picking efficiency. In order to complete the intensive picking work as soon as possible during the apple ripening period, the fruit picking robot not only needs to be able to work normally during the day but also needs to have the ability to work at night. However, due to insufficient light at night, it is difficult for the vision system of the picking robot to obtain a clear image, making the robot unable to work normally. [0003] In 1997, Sun Ming et al. analyzed the hi...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies in the prior art, the present invention provides a night recognition method for an apple picking robot, which solves the problems of selection and arrangement of light sources in the night recognition of an apple picking robot and the problems of shadows and reflections in nighttime images, making the apple picking robot Works fine at night

Images