Garbage identification and classification method, computer readable storage medium and robot

Abstract

The invention discloses a garbage identification and classification method, a computer readable storage medium and a robot, and the method comprises the steps: carrying out the deblurring of an original frame image according to a preset deblurring network, and obtaining a target image; inputting the target image into a preset deep convolutional neural network model for target object detection to obtain a first feature vector diagram of the target object; and tracking the target object according to the first feature vector diagram, and sorting the target object according to the category and the position of the target object. According to the recognition method, the false detection rate and the omission ratio of ground garbage recognition can be effectively reduced, the detection speed and universality are improved, and the recognition method is less affected by the environment.

Description

technical field [0001] The invention relates to the technical field of robots, in particular to a garbage identification and classification method, a computer-readable storage medium and a robot. Background technique [0002] In addition to cleaning fallen leaves, paper scraps and other debris during garbage inspection, traditional robots also need to detect, classify and collect recyclable garbage, so as to realize the recycling of recyclable garbage. When traditional algorithms detect recyclable garbage, they mostly segment the road surface in the image, and use algorithms such as connected domain detection, grayscale histogram, and feature matching to judge the category and location information of objects on the road surface, and then identify specific recyclable garbage. Garbage, such as plastic bottles, cans, etc., for recycling. However, because most of the garbage on the road has been squeezed and has different shapes, the feature matching effect is mostly poor. In ...

AI Technical Summary

Problems solved by technology

However, because most of the garbage on the road has been squeezed and has different shapes, the feature matching effect is mostly poor.

In addition, due to the change of the viewing angle position during the movement of the robot, the shape and size of the same object will change, and it is easy to miss multiple frames in several consecutive frames. At this time, the robot will update the path and point downward One garbage march, resulting in inefficient operations

Images