Waste mobile phone model identification method based on convolutional neural network

Abstract

The invention provides a waste mobile phone model identification method based on a convolutional neural network in order to solve the problem that models are difficult to identify accurately in the waste mobile phone recovery process. According to the method, an edge detection algorithm is used to analyze regional features in a mobile phone verification photo, a weight-sharing feature extraction convolutional network is constructed, similarity between waste mobile phone image regional features and a standard sample is evaluated, rapid identification of a mobile phone model is realized. According to the method, good rapidity and accuracy of mobile phone model identification in different scenes are kept, and the waste mobile phone recovery efficiency and the economic benefits of recovery enterprises can be improved.

Description

technical field [0001] The invention utilizes a low-rank convolutional neural network-based identification method for used mobile phone models to realize accurate identification of mobile phone models in the recycling process of used mobile phones. In the process of recycling waste mobile phones, it is possible to obtain greater economic benefits by classifying mobile phones by model. The identification of mobile phone models has become an important factor affecting the efficiency of waste mobile phone recycling. There are many brands and models of mobile phones with high similarity, so it is necessary to have a certain Only with the accumulation of experience can we skillfully distinguish between mobile phone models. Applying the convolutional neural network-based model identification method of used mobile phones to the recycling process of used mobile phones can avoid problems such as classification errors and low classification efficiency caused by inexperienced personnel, ...

AI Technical Summary

Problems solved by technology

[0006] The present invention obtains a low-rank bilinear convolutional neural network mobile phone model identification method. The identification method extracts the recognizable mobile phone area in the inspection photo through the low-rank convolutional algorithm, and uses the convolutional neural network to realize the recognition of waste Fast and accurate identification of mobile phone models; solves the problem of model identification in the recycling process of waste mobile phones, and improves the recycling efficiency of mobile phones

Images