首页> 外文会议>International Conference on Computer, Communication and Signal Processing >Rainy Image Enhancement Using Modified Multistage Gaussian Filter and Weighted Median Guided Filter
【24h】

Rainy Image Enhancement Using Modified Multistage Gaussian Filter and Weighted Median Guided Filter

机译:使用改进的多级高斯滤波器和加权中值引导滤波器的多雨图像增强

获取原文

摘要

Rain streaks and droplets removal from an image is quite complex. In rainy days, the performance of outdoor vision system will be degraded due to poor visibility, deformation and blurring caused by rain. In this paper, modified multi-stage Gaussian filtering technique for rain streaks removal and a method for removing rain droplets on windscreen are proposed. Rain removed image is processed using dark channel prior algorithm for haze removal. Dark channel prior is combined with triple clipped dynamic histogram equalization technique for sharpness enhancement. The proposed algorithm is compared with that of Shi et al. [6] in terms of peak signal -to- noise ratio (PSNR) and structural similarity index (SSIM). Simulation results on variety of rain streaks and rain droplets images yielded an average PSNR and SSIM values of 52dB and 0.99 respectively using the proposed algorithm as compared with 32dB and 0.96 obtained using algorithm [6].
机译:从图像中移除雨条纹和液滴非常复杂。在雨天,由于雨引起的可见性,变形和模糊,户外视觉系统的性能将降低。本文提出了一种改进的多级高斯滤波技术,用于去除雨条纹和去除挡风玻璃上雨滴的方法。使用用于雾化的黑色通道现有算法处理雨量移除图像。黑暗通道先前与三重剪裁动态直方图均衡技术相结合,用于清晰度增强。该算法与Shi等人的算法进行了比较。 [6]就峰值信号 - 噪声比(PSNR)和结构相似度指数(SSIM)而言。与使用算法获得的32dB和0.96相比,各种雨条纹和雨液滴图像的仿真结果分别使用所提出的算法分别产生52dB和0.99的平均PSNR和SSIM值。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号