Fruit picking robot target detection method based on deep learning in unstructured environment

Abstract

The invention relates to a fruit picking robot target detection method based on deep learning in an unstructured environment, and belongs to the technical field of intelligent agricultural production.According to the method, a Mask R-CNN is used as a target detection framework, ResNet-101 is used as a backbone network and is combined with an FPN architecture to perform target feature extraction,then a feature map output by the backbone network is sent to an RPN to generate RoI, and then the RoI output from the RPN is mapped to extract corresponding target features in a shared feature map; and finally, the features are respectively output to an FC layer and an FCN layer to carry out target detection, frame regression and instance segmentation. According to the method, the problem of low detection precision caused by illumination condition change, branch and leaf shielding, fruit clustering overlapping and the like of a traditional digital image processing technology in an unstructuredenvironment is solved, and the defects of complex structure, slow gradient disappearance, large calculation training amount, slow model convergence and the like of a common target detection neural network are also overcome.

Description

technical field [0001] The invention belongs to the technical field of intelligent agricultural production, and relates to a target recognition and detection method for a fruit and vegetable picking robot, in particular to a target detection method for a fruit picking robot based on deep learning in an unstructured environment. Background technique [0002] Since the advent of fruit and vegetable picking robots, many new methods have emerged on target detection and recognition. Traditional digital image processing technology, such as feature extraction and recognition based on color, texture, shape, etc., has the disadvantage of relying on the external color, texture, and shape of the target fruit, but the external features of the fruit will change with the different growth stages of the fruit , and is easily affected by lighting conditions or the inability to obtain complete external characteristics due to shading by branches and leaves and overlapping fruits. [0003] Wit...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the object of the present invention is to provide a target detection method for fruit picking robots based on deep learning in an unstructured environment, to solve the problem that traditional target detection is easily affected by lighting conditions, geometric features change with the growth cycle, branches and leaves block fruit and other problems. Problems such as fruit cluster growth have also overcome the shortcomings of general deep learning neural network models that require a large number of training sets, a large amount of calculation, long calculation time, high requirements for computer hardware, and insufficient recognition accuracy.

Images