基于局部自适应对比度增强算法的木板条纹识别  

Wood Stripe Recognition Based on Local Adaptive Contrast Enhancement Algorithm

作　　者：虞成俊 彭文刚 YU Chengjun;PENG Wengang(School of Mechanics and Power,China Three Gorges University,Yichang Hubei 443002,China)

机构地区：[1]三峡大学机械与动力学院,湖北宜昌443002

出　　处：《信息与电脑》2020年第22期57-59,共3页Information & Computer

摘　　要：为了满足对木板条纹识别准确性与速度性的要求,针对由于光照不均或摄像头硬件清晰度达不到要求而导致的木板图像对比度差,不利于人工与机器识别的现象,本文采用基于局部自适应对比度增强(ACE)的算法对采集到的木板图像进行图像增强.首先,对原图像进行gamma校正使图像像素集中区域得到扩展;其次,采用低通滤波得到图像的低频图像,通过原图减去低频图像从而得到图像中的高频部分,接着对高频部分进行增益,并与低频部分合并从而得到增强后的图像;最后,通过双边滤波达到去噪作用,且结合均方误差(MSE)、结构相似性(SSIM)及平均梯度法(Average gradient)等评价指标,将本文算法与直方图均衡化(HE)和多尺度Ret inex(MSR)对比.结果分析表明,该算法在提高对比度以及保留图像细节等方面表现出可靠的性能.In order to meet the requirements of accuracy and speed of wood stripe recognition,aiming at the phenomenon of poor contrast of wood plate image caused by uneven illumination or camera hardware definition,which is not conducive to manual and machine recognition,this paper adopts an algorithm based on local adaptive contrast enhancement(ACE)to enhance the image of the collected board.Firstly,gamma correction is applied to the original image to expand the pixel concentration area of the image;secondly,the low-pass filter is used to obtain the low-frequency image of the image,and the high-frequency part of the image is obtained by subtracting the low-frequency image from the original image,and then the high-frequency part is added and combined with the low-frequency part to obtain the enhanced image;finally,the bilateral filtering is used to achieve the denoising effect,Combined with MSE,SSIM and average gradient,the algorithm is compared with histogram equalization(he)and multi-scale Retinex(MSR).The results show that the algorithm has reliable performance in improving contrast and preserving image details.

关 键 词：机器识别 图像增强 木板条纹 ACE算法 双边滤波 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]