摘要
图像在拍摄、传输过程中常受到设备和环境噪声的干扰,噪声种类和滤波方法也很多,在实际应用中很难选择最合适的滤波器,而图像质量评价法检测的是两幅图像的相似度,不能反映滤波器的滤波性能。针对这一问题,根据图像质量评价标准,提出抗噪参数P的图像滤波性能检测,该方法首先采用几种滤波器对原图和加噪图像进行滤波,求取滤波后图像的PSNR值,每种滤波器会对应两个PSNR值,分别取有无噪声时的PSNR值的绝对差值,然后将绝对差值比上无噪声时的PSNR值即为滤波器的抗噪性能。通过大量实验图像对比和实验数据分析,本检测方法对滤波器性能的检测具有可行性。
Image is often disturbed by the noise of equipment and environment in the process of shooting and transmitting.Many kinds of noise and filtering method makes it difficult to select the most appropriate filter in practical application.The image quality evaluation method is to evaluate the similarity of two images,which can not reflect the filter performance.According to the image quality evaluation standard,the image filtering performance of Pis proposed in the paper.Firstly,The method filters the original image and noise image by several filters to get thethe peak signal-to-noise ratio PSNR of original image to the original image filtering and noisy image to noise filtering image.Each filter corresponds to two PSNR values,obtaining the absolute difference of the PSNR value with no noise and noisy image.Then the absolute difference is divided by the PSNR value of the non noise that is the anti-noise performance of the filter.A large number of experiments on image and experimental data analysis prove that this method on filter performance detection is feasible.
出处
《湖北工业大学学报》
2016年第5期43-47,共5页
Journal of Hubei University of Technology
关键词
图像质量评价
滤波性能
抗噪参数
峰值性噪比
噪声图像
image quality assessment
filtering performance
anti-noise parameters
peak signal-to-noise ratio
noisy image
