基于变分自编码器的输送带煤量分级算法研究

Research on Classification Algorithm of Belt Coal Quantity Based on Variational Autoencoder

为了能够高效精准地对煤矿带式输送机上的煤量进行分级,提出了一种基于变分自编码器(VAE)的输送带煤量分级算法。首先为了解决真实场景图像往往存在许多噪声信息的问题,利用VAE对图像进行重构处理,使图像更加光滑以减少噪声信息对后续分级的干扰。然后为了提升输送带煤量的分级精度,利用卷积神经网络(CNN)对重构后的图像进行分级预测。实验结果表明,相对于对比方法,此该算法在各评价指标上均有提升,同时重构图像能够保留原始图像的关键信息。

In order to efficiently and accurately classify the coal quantity on the coal mine belt conveyor, a belt coal quantity classification algorithm based on variational autoencoder(VAE) was proposed. First of all, in order to solve the problem that real scene images often have a lot of noise information, VAE was used to reconstruct the image to make the image smoother, so as to reduce the interference of noise information on subsequent grading. Then in order to improve the classification accuracy of the belt coal quantity, a convolutional neural network(CNN) was used to classify and predict the reconstructed image. The experimental results show that compared with the comparison method, this algorithm has improved various evaluation indexes, and the reconstructed image can retain the key information of the original image.