摘要
为准确识别输送带纵向撕裂,将自组织特征映射(SOM)引入到输送带纵向撕裂检测中。输送带图像经过中值滤波预处理之后,提取其梯度方向直方图特征作为SOM网络的输入,采用卡方距离描述特征之间的相似性,建立了输送带图像纵向撕裂的SOM检测模型,详细介绍SOM训练过程,最后对文中算法进行了仿真实验。实验结果表明采用SOM网络识别输送带纵向撕裂具有良好的效果,为输送带自动识别拓宽了思路。
To recognize longitudinal tearing accurately, self - organizing feature maps （SOM） was introduced to detect the longitudinal tearing of conveyor belt. After median filtering, the histograms of oriented gradient of the conveyor belt images were extracted as the SOM＇s input feature vectors. Then Chi - square distance was adopted to describe similarity of two different feature vectors. SOM detecting model was built and training process of the model was introduced in detail. Simulation experiment was conducted to test the proposed algorithm. The result shows that, the SOM network works well in recognizing longitudinal tearing of conveyor belt.
出处
《煤炭工程》
北大核心
2015年第9期114-116,共3页
Coal Engineering
基金
中央高校基本科研业务费专项资金(ZY20140216)
关键词
输送带
纵向撕裂
自组织特征映射
神经网络
conveyor belt
longitudinal tearing
self - organizing feature maps
neural network
