融合眼视觉和改进多尺度Retinex的矿岩裂隙图像增强  

Enchantment of Ore-rock Fissure Images Based on Human Visual Perception and Improved Multi-scale Retinex Algorithm

作　　者：赵杰[1] 汪洪法 周明 ZHAO Jie;WANG Hongfa;ZHOU Ming(Department of Electronic Information,Jinzhong Vocational and Technical College,Jinzhong 030600,China;School of Information and Computer Science,Taiyuan University of Technology,Taiyuan 030024,China)

机构地区：[1]晋中职业技术学院电子信息系,山西晋中030600 [2]太原理工大学信息与计算机学院,山西太原030024

出　　处：《金属矿山》2024年第12期252-257,共6页Metal Mine

基　　金：山西省教育科学“十四五”规划基金项目(编号:GH-220913)。

摘　　要：准确有效地检测出矿岩裂隙对于实现矿岩体稳定性分析具有重要作用。针对深度网络对低光照条件下采集的低分辨率图像存在特征提取不充分的问题,提出了一种融合人眼视觉和改进多尺度Retinex的矿岩裂隙图像增强算法。首先,结合人眼视觉特征和小波变换算法对采集的矿岩裂隙低分辨图像进行预处理;然后,采用改进的多尺度Retinex算法从光照分量和反射分量方面估计噪声,并排除其对图像的干扰;最后,利用CBAM (Convolutional block attention module)注意力机制强化去噪后的图像在通道和空间维度的表达能力,实现矿岩裂隙图像低分辨图像增强。试验结果表明:与传统的Retinex算法相比,改进的多尺度Retinex模型能够更好地提取图像细节信息,并且在保持自然观感的同时,增强了图像对比度,使得裂隙更加清晰可见。Accurate and effective detection of ore-rock fissure plays an important role in the stability analysis of rock mass.Aiming at the problem of insufficient feature extraction in low resolution images acquired under low light conditions by deep network,a new algorithm for ore-rock fissure image enhancement based on human vision and improved multi-scale Retinex is proposed.Firstly,the low-resolution images of ore-rock fissure are preprocessed by combining human visual features and wavelet transform algorithm.Then,the improved multi-scale Retinex algorithm is used to estimate the noise from illumination and reflection components,and eliminate the interference to the image.Finally,CBAM(Convolutional Block Attention Module)is used to enhance the expression ability of denoised images in channel and spatial dimensions,and to achieve low-resolution image enhancement of mine and rock fissure images.The experimental results show that compared with the traditional Retinex algorithm,the improved multi-scale Retinex model can better extract image details,enhance image contrast and make cracks more clearly visible while maintaining the natural perception.

关 键 词：矿岩裂隙 图像增强 多尺度RETINEX算法 图像去噪 小波变换 CBAM注意力机制 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]