Method for carrying out garbage category detection by using neural network under fixed distance

Abstract

The invention discloses a method for carrying out garbage category detection by using a neural network under a fixed distance. The position of a camera device which is arranged based on each throwing point and is used for shooting garbage pictures is fixed, the classification detection of garbage is always kept at a fixed distance, the size of the corresponding garbage also has a certain range in the shot garbage picture, when the detection classification layer performs detection and judgment, after the target frames are determined, the target frames are screened, the screened target frames are classified according to large targets and small targets, and then the six pictures are divided into two groups of large targets and small targets to be processed respectively, so the classification judgment process of the SSD target detection neural network based on deep learning is greatly simplified, and the speed and precision of the SSD target detection neural network based on deep learning are effectively improved. The method has the advantages of high detection speed and high detection precision.

Description

technical field [0001] The invention relates to a method for detecting garbage categories by using a neural network, in particular to a method for detecting garbage categories by using a neural network at a fixed distance. Background technique [0002] With the improvement of people's living standards, more and more garbage is produced. At present, these garbage are mainly divided into recyclable garbage, harmful garbage, kitchen waste and other garbage. By classifying and managing waste, the recycling of waste resources can be realized to the greatest extent while reducing the amount of waste to be disposed of. [0003] At present, neural networks are usually used for garbage category detection, and the accuracy and real-time performance of neural network models directly determine the accuracy and real-time performance of garbage category detection. SSD target detection neural network based on deep learning is a commonly used neural network currently used for garbage class...

AI Technical Summary

Problems solved by technology

[0006] However, when faced with a large number of dense garbage distribution pictures, the above-mentioned SSD target detection neural network based on deep learning needs to process the garbage one by one because there are too many garbage in the picture, and the shapes and sizes of the garbage are different, so that the garbage The speed and accuracy of category detection are not high