Material identification method, device and equipment and storage medium

Abstract

The embodiment of the invention discloses a material identification method, device and equipment and a storage medium. The method comprises the steps: obtaining a to-be-identified material image; processing the to-be-recognized material image by adopting a set edge recognition algorithm to obtain local feature information; inputting the to-be-identified material image into a convolutional layer ina set neural network to obtain depth feature information, wherein the set neural network comprises a convolution layer and a full connection layer; performing weighted fusion on the local feature information and the depth feature information to obtain intermediate features; and inputting the intermediate features into a full connection layer of a set neural network to obtain the material type ofthe to-be-identified material. According to the technical scheme provided by the embodiment of the invention, the problem of identification failure caused by death and incapability of reactivating part of neurons in the neural network when only the set neural network is adopted for material identification is solved, the accuracy of material identification is improved, and meanwhile, the error-tolerant rate of the neural network when the set neural network is adopted for material identification is improved.

Description

technical field [0001] The embodiments of the present invention relate to the technical field of construction machinery, and in particular to a material identification method, device, equipment and storage medium. Background technique [0002] With the continuous advancement of science and technology, people's pursuit of comfort, safety and efficiency in work and life has become more and more obvious, and intelligence and even artificial intelligence have become an important means to achieve this goal. Due to the harsh working environment during the earthwork, there are various problems such as high temperature, dust, humidity, and being far away from the living area, the demand for unmanned is becoming stronger and stronger. [0003] For the identification of the types of materials collected in earthmoving machinery, at present, the key characteristic information of materials is usually obtained through sound, light and other media for identification. Most of them use deep ...

AI Technical Summary

Problems solved by technology

[0004] However, when traditional machine vision is used to identify materials, the identification process is often slow due to the need for multiple treatments on the materials during the identification process, and the same treatment will produce different processing results for different materials, resulting in lower recognition accuracy. Low

However, when the material is identified through the Alexnet network model, the network model will have the phenomenon that the neural unit dies and cannot be revived in some places in the obtained material image, which reduces the accuracy of material recognition

Images