基于NSCT和改进Pal_King算法的电力设备红外图像增强方法  

INFRARED IMAGE ENHANCEMENT METHOD FOR POWER EQUIPMENT BASED ON NSCT AND IMPROVED PAL_KING ALGORITHM

作　　者：陈尧 尹丽菊[1] 咸日常[1] 潘金凤[1] 王季峥 于毅 Chen Yao;Yin Liju;Xian Richang;Pan Jinfeng;Wang Jizheng;Yu Yi(School of Electrical and Electronic Engineering,Shandong University of Technology,Zibo 255000,Shandong,China)

机构地区：[1]山东理工大学电气与电子工程学院,山东淄博255000

出　　处：《计算机应用与软件》2021年第11期236-241,287,共7页Computer Applications and Software

摘　　要：针对电力设备红外图像存在对比度较低、噪声复杂、清晰度不优的问题,提出一种基于NSCT和改进PalKing算法的电力设备红外图像增强方法。采用直方图双向均衡化对红外测温图像进行预处理,对NSCT变换产生的低高频子带分别采用线性增强和改进的自适应阈值函数进行系数修正,通过NSCT反变换重构图像,利用改进的PalKing算法进行图像增强处理。实验结果表明,该方法输出图像的平均梯度值、峰值信噪比值和熵值分别保持在5.9、25.4和6.8以上,提高了电力设备红外图像的对比度,有效抑制了噪声,使图像更加清晰。Aimed at the problem that the infrared image of power equipment has the problem of low contrast, complex noise and poor definition, an infrared image enhancement method based on NSCT and improved PalKing algorithm is proposed. The method preprocessed the infrared temperature measurement image by histogram bidirectional equalization, and then used the linear enhancement and the improved adaptive threshold function to correct the coefficients of the low-frequency sub-bands generated by the NSCT transform, and the image was reconstructed by NSCT inverse transform. The image was constructed, and the image enhancement processing was performed using the improved PalKing algorithm. The experimental results show that the average gradient value, peak signal-to-noise ratio and entropy of the output image of the method are kept above 5.9, 25.4 and 6.8, respectively, which improves the contrast of the infrared image of the power equipment, effectively suppresses the noise and makes the image clearer.

关 键 词：图像增强 NSCT变换 Pal_King 自适应阈值 电力设备 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]