输送带故障检测多视点图像自适应增强方法

为提高矿用输送带故障检测的准确性,针对输送带故障检测采集的图像存在光照不均、图像质量差、影响故障检测与识别的问题,提出输送带故障检测多视点图像自适应增强方法。采用具有尺度自适应功能的多尺度Retinex法对图像光照分量进行提取,再根据提取的光照分量,利用针对多视点线阵相机采集图像特性改进的自适应伽马函数,完成对多视点图像的非均匀光照校正,最后进行图像细节对比度的增强。实验结果表明:该方法可去除非均匀光照、井下恶劣环境等对图像质量的影响,使不同视点不同光照环境条件下线阵相机采集图像的平均亮度具有一致性,压缩图像动态范围,提高对比度,使得图像中阴暗区域的细节更加明显,提高故障检测准确性,适合于矿用输送带故障在线检测。

Multi-view image adaptive enhancement method for conveyor belt fault detection

The images collected to detect the fault of conveyor belt have some problems of non-uniform illumination,and poor image quality.It makes fault detection and recognition become difficult.In order to improve the accuracy of fault detection of mine conveyor belt,a method for adaptive enhancement of multi-view images has been proposed in this paper.Firstly,a multi-scale Retinex method with scale adaptive is used to extract the illumination components of an image.According to the extracted illumination component,the non-uniform illumination correction of multi-view images is achieved by using the adaptive gamma function.Finally,the image contrast is enhanced.The experimental results show that this method has many effects,such as removing the effect of non-uniform illumination and coal mine harsh environments,making the average intensity of the image collected by the linear array camera under different viewpoints and illumination conditions consistent,compressing the image dynamic range,improving the image contrast,making the detail of dark area more obvious in the image.This method contributes to improve the accuracy of fault detection,and is suitable for the on-line fault detection of mine conveyor belts.