改进的提升小波域内的图像融合算法在全景安检系统中的应用  

Image Fusion Algorithm Applied to Panoramic Security Check in Lmproved Lifting Wavelet Transform Domain

作　　者：王帅[1] 邹佳彬 张莹莹[1] 张江州[1] 闫淮海 WANG Shuai ZOU Jia-bin ZHANG Ying-ying ZHANG Jiang-zhou YAN Huai-hai(Information Research Institute, Shandong Academy of Sciences, Jinan 250014, China School of Information and Electrical Engineering, China University of Mining ＆ Technology, Xuzhou 221008, China)

机构地区：[1]山东省科学院情报研究所,济南250014 [2]中国矿业大学信息与电气工程学院,徐州221008

出　　处：《计算机科学》2016年第B12期156-161,共6页Computer Science

基　　金：本文受山东省重点研发计划项目（2015GSFl20009）资助.

摘　　要：针对智能车辆安检系统图像融合处理的问题，采用一种同时对图像行列进行提升的小波变换方法，在该改进的提升小波变换域中对得到的近似系数与细节系数采用不同的融合方式。在选择近似系数时，采用基于局部方向信息的特征对比度系数选择方案。在选择细节系数时，采用基于方向信息测度多尺度积的特征对比度的系数选择方案。实验结果表明，该算法相对传统基于小波变换的融合算法，可以有效抑制图像中存在的噪声影响，能得到更好的融合效果。Aiming at the problem of intelligent vehicle security system in image fusion processing, an improved lifting wavelet transform method which can deal with procession of image at the same time was used. In improved lifting wavelet transform domain, different fusion methods for approximate coefficients and detail coefficients was chosen. When choosing approximation coefficients, a new coefficients options selection was used based on local orientation information feature contrast and visual characteristic contrast. When choosing detail coefficients, a new coefficients options selection was used based on the local orientation information feature contrast, multiscale products and characteristics contrast. The experimental results show that the algorithm is relative to the traditional fusion algorithm based on wavelet transform. It can effectively suppress the influence of the noise in the image and get better fusion effect.

关 键 词：图像融合 改进的提升小波变换 方向信息测度 多尺度积 特征对比度 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]