Asphalt concrete detection system and method based on fog generative adversarial neural network

Abstract

The invention discloses an asphalt concrete detection system and method based on fog generation confrontation neural network, comprising a computer system, an image acquisition device, etc. The image acquisition device acquires an image of asphalt concrete after mixing, and transmits the image to a computer system to realize online detection Asphalt concrete quality; the present invention combines machine vision and neural network algorithm, performs de-smog and de-fuzzification operations on asphalt images through fog generation confrontation neural network model, and then uses convolutional neural network to detect the uniformity of the appearance of asphalt concrete, and judges the quality of asphalt concrete. Whether there is white material in asphalt concrete, whether there is agglomeration phenomenon and serious segregation phenomenon, calculate in real time whether the quality of asphalt concrete is qualified. The invention has high degree of automation, high accuracy, good real-time performance and high efficiency, can form a complete monitoring system with the mixing station system, can timely discover the quality problems existing in the mixing of asphalt concrete, effectively reduce economic losses and save energy time.

Description

technical field [0001] The invention belongs to the technical field of machine vision defect detection, and in particular relates to an asphalt concrete detection system and method based on a fog generation confrontation neural network. Background technique [0002] Asphalt pavement is more durable than traditional cement pavement, has strong water permeability and high wear resistance, and is the first choice for pavement construction. With the improvement of the technical level in urban road construction, the quality requirements for asphalt concrete are getting higher and higher. The mixing quality of asphalt concrete materials is a key step in the construction process. If the mixing quality is not up to standard, it will directly affect the asphalt concrete structure. adhesion, stability and safety. In the mixing process of asphalt concrete, it is necessary to ensure uniform mixing, no whitening, no agglomeration or serious segregation. The current method for judging t...

AI Technical Summary

Problems solved by technology

At present, the method of judging the quality of asphalt concrete is to manually observe the appearance of asphalt concrete at the position of the receiving channel of the finished material outlet, which has problems such as hysteresis and low detection efficiency. serious threat

Images