High-noise environment kiln car identifier identification method and system based on convolutional neural network

Abstract

The invention discloses a high-noise environment kiln car identifier identification method and system based on a convolutional neural network. The identification method is an edge detection algorithm-edge enhancement type edge detection algorithm adopting a Canny edge detection algorithm, the Gaussian smoothing step of Gaussian blur of an image in the algorithm is to construct a Gaussian convolution kernel to perform convolution processing on the image, and when a pixel matrix does not contain boundary pixel information, a Gaussian weight coefficient matrix is established by taking weight coefficients from coordinate positions, and when boundary information is contained, the weight coefficients corresponding to boundary pixels are changed to form an improved Gaussian weight coefficient matrix. By means of the technical scheme, the method and system are applied to a wall material production line, signboards on kiln cars can be identified, and the position of each kiln car and whether the state of each kiln car is accurate or not can be dynamically monitored through configuration software. The production efficiency can be improved and the production reliability can be enhanced while manpower is replaced.

Description

technical field [0001] The invention relates to the field of image recognition, in particular to a kiln car mark recognition method in a high-noise environment based on a convolutional neural network. Background technique [0002] At present, in domestic brick and tile sintering factories, tunnel kiln has become the preferred brick and tile production equipment, which is an important symbol of the rapid development and progress of my country's brick and tile industry technology. However, at present, the automation level of domestic tunnel kiln production process is generally relatively backward. The kiln cars of the existing wall material manufacturing production line still need to be manually operated, and the production line equipment is still in the manual control stage, relying on manpower to count the number of kiln cars and the number of kiln cars. state and position, so it has the following disadvantages: [0003] 1. Waste of human resources; [0004] 2. Hinder the ...

AI Technical Summary

Problems solved by technology

[0003] 1. Waste of human resources;

[0004] 2. Hinder the development from semi-automation to automation and intelligence;

[0005] 3. Lack of intelligence

[0006] Therefore, there is an urgent need for automatic transformation of this equipment, which must involve target identification, usually using the method of setting signs, and for now, my country's sign recognition technology is in a stage of rapid development, which is specifically reflected in processing accuracy, reproducibility, flexibility, However, in the actual development process, the development of this technology is still limited by actual needs

Images