摘要
针对当前图像画质增强算法存在的亮度低、噪声大、对比度小等缺陷,为改善传感器输出图像的视觉效果,设计基于图像传感器的图像画质增强算法。首先,采用图像传感器对目标图像进行采集,并引入双边滤波算法对采集的原始图像进行去噪,抑制噪声图像画质干扰;然后,采用幂次变换提高去噪图像的对比度,并采用非线性变换对去噪图像进行亮度增强;最后,采用Retinex算法根据人眼视觉感知特性对图像整体进行增强,使图像的不同区域过渡更加自然,并进行了图像画质增强性能的验证性测试实验。结果表明,所提算法较好地解决了当前图像画质增强过程中存在的一些难题,提高了图像的信噪比,亮度和对比度也得到了明显的改善,相对于对比算法,获得了十分理想的图像画质增强结果。
For the defects of the current image quality enhancement algorithms,such as low brightness,high noise and low contrast,an image quality enhancement algorithm based on image sensor is designed to improve the visual effect of the sensor output image.Firstly,the target image is collected with the image sensor,and the collected original image is denoised by bilateral filtering algorithm to suppress the interference of noise on image quality.Then the contrast of the denoised image is improved by power transform,and the brightness of the denoised image is enhanced by non-linear transform.Finally,the whole image is enhanced by Retinex algorithm according to the human visual perception characteristics,so that the transition of different areas in the image is more natural.A test experiment was performed to verify the enhancement performance of image quality.The results show that the proposed algorithm can deal with certain difficulties existing in the current image quality enhancement process,improve the signal-to-noise ratio of the image,and obviously enhance the brightness and the contrast of the image.In comparison with the existing algorithms,the proposed algorithm can obtain a satisfied enhancement result of image quality.
作者
刘纯
陶薇薇
LIU Chun;TAO Weiwei(School of Software,Chongqing Institute of Engineering,Chongqing 400000,China)
出处
《现代电子技术》
北大核心
2020年第7期66-69,共4页
Modern Electronics Technique
基金
重庆市高等教育教学改革研究重大项目(171028)
重庆市教育科学规划项目(2017 GX 364)。
关键词
画质增强算法
图像处理技术
图像去噪
信号采集
双边滤波
性能测试
image quality enhancement algorithm
image processing technology
image denoising
signal acquisition
bilateral filtering
performance testing
作者简介
刘纯(1987—),女,重庆云阳人,硕士,高级工程师,研究方向为数字媒体;陶薇薇(1982—),女,土家族,重庆秀山人,硕士,高级工程师,研究方向为数字媒体。
