直方图均衡化下图像模糊边缘轮廓提取仿真  

Simulation of Image Fuzzy Edge Contour Extraction under Histogram Equalization

作　　者：贾俣 梅芳[2] JIA Yu;MEI Fang(Department of Investigation,Fujian Police College,Fuzhou Fujian 350007,China;School of Computer and Information,Jiangxi Agricultural University,Nanchang Jiangxi 330045,China)

机构地区：[1]福建警察学院侦查系,福建福州350007 [2]江西农业大学计算机与信息学院,江西南昌330045

出　　处：《计算机仿真》2025年第2期395-398,404,共5页Computer Simulation

摘　　要：受不同光照条件以及采集设备性能影响,图像存在较多噪声,其边缘信息被模糊化,丧失了细节和清晰度,导致边缘的定位和提取较为复杂。为此,提出直方图均衡化下图像模糊边缘轮廓提取方法。采用平稳小波变换方法分解图像,归一处理相关值,标记图像中的边缘与噪声点,剔除噪声点;利用直方图均衡化处理去噪后的图像,增强图像中存在的细节信息;利用引导滤波算法增强图像边缘,通过OTSU算法确定图像分割的最佳阈值,获得图像的边缘轮廓。仿真结果表明,所提方法可有效消除图像中存在的噪声、图像清晰度高、轮廓提取精度高,平均相对误差仅为0.02%。At present,images generally contain many noises due to different lighting conditions and the performance of acquisition equipment.And their edge information is blurred,losing details and clarity,so the positioning and extraction of edge becomes more complex.Therefore,this paper presented a method of extracting fuzzy edge contour of image under histogram equalization.Firstly,we adopted the stationary wavelet transform method to decompose the image,and normalized the related values.After marking the edges and noise points in the image,we eliminated these noise points.Then,we used histogram equalization to process the denoised image,thereby enhancing the details in the image.Next,we used the guided filter algorithm to enhance the image edge.Through the OTSU algorithm,we determined the optimal threshold value for image segmentation.Finally,we obtained the edge contour of image.Simulation results prove that the proposed method can effectively eliminate the noise,and has high clarity of image as well as high precision of contour extraction.The mean error is only 0.02%.

关 键 词：图像边缘轮廓 直方图均衡化 平稳小波变换 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]