测绘通报 ›› 2018, Vol. 0 ›› Issue (4): 28-31.doi: 10.13474/j.cnki.11-2246.2018.0105

• 行业观察 • 上一篇    下一篇

小波阈值改进算法的遥感图像去噪

陈竹安1,2,3,4, 胡志峰1   

  1. 1. 东华理工大学测绘工程学院, 江西 南昌 330013;
    2. 流域生态与地理环境 监测国家测绘地理信息局重点实验室, 江西 南昌 330013;
    3. 江西省数字国土重点实验室, 江西 南昌 330013;
    4. 江西生态文明建设制度研究中心, 江西 南昌 330013
  • 收稿日期:2017-07-26 修回日期:2017-09-10 出版日期:2018-04-25 发布日期:2018-05-03
  • 通讯作者: 胡志峰。E-mail:291151928@qq.com E-mail:291151928@qq.com
  • 作者简介:陈竹安(1978-),男,硕士,副教授,研究方向为测绘技术、土地信息技术、遥感数据处理。E-mail:zachen@ecit.cn
  • 基金资助:

    国家自然科学基金(51708098);江西省教育厅课题(GJJ160537);江西省高校人文社会科学课题(GL1501);流域生态与地理环境监测国家测绘地理信息局重点实验室课题(WE2016018);江西省数字国土重点实验室开放研究基金项目(DLLJ201720)

Remote Sensing Image Denoising Based on Improved Wavelet Threshold Algorithm

CHEN Zhu'an1,2,3,4, HU Zhifeng1   

  1. 1. Faculty of Geomatics, East China University of Technology, Nanchang 330013, China;
    2. Key Laboratory of Watershed Ecology and Geographical Environment Monitoring, NASG, Nanchang 330013, China;
    3. Jiangxi Province Key Laboratory of Digital Land, Nanchang 330013, China;
    4. Jiangxi Ecological Civilization Construction System Research Center, Nanchang 330013, China
  • Received:2017-07-26 Revised:2017-09-10 Online:2018-04-25 Published:2018-05-03

摘要:

通过对于文献中已有小波阈值去噪方法的研究,结合已有的一些小波阈值去噪函数,提出了相应的小波阈值去噪的改进方法来完善和提高小波阈值去噪的处理能力和可行性。该阈值函数加入了有效的调整系数来控制函数的可变。该函数不但同时保留了相应的传统小波硬阈值、软阈值衍化的优点,也提高了相应精度指标。利用该函数阈值去噪不仅在经典的图像中起到很大的改善,在遥感图像的去噪处理方面也有明显的精度提高。该方法通过去噪评价指标均方差(MSE),峰值信噪比(PSNR),信噪比(SNR),均方根误差(RMSE)进行去噪后图像的评价。该改进的阈值函数方法对于图像的处理后评价指标明显有所改善。

关键词: 小波去噪, 遥感图像, 改进阈值函数, 评价方法

Abstract:

Based on the research of wavelet threshold denoising method in the literature, an improved method of wavelet threshold denoising is proposed to improve and improve the processing capability and feasibility of wavelet threshold denoising. Combining some of the existing wavelet threshold denoising functions. The threshold function adds a valid adjustment factor to control the variable function. This function not only preserves the corresponding advantages of traditional wavelet hard threshold and soft threshold, but also improves the corresponding precision index. Using this function, threshold denoising not only improves greatly in classical images, but also improves the accuracy of noise detection in remote sensing images. The method evaluates the image by denoising after denoising evaluation mean square error (MSE), peak signal to noise ratio (PSNR), signal to noise ratio (SNR), and root mean square error (RMSE). The improved threshold function method has obviously improved the post-treatment evaluation index for the image.

Key words: wavelet denoising, remote sensing image, improved threshold function, evaluation method

中图分类号: