矿山遥感图像小波域模糊隶属度阈值去噪算法

发布时间：2018-03-29 15:40

本文选题：矿山测量　切入点：矿山遥感图像　出处：《金属矿山》2017年04期

【摘要】：矿山遥感图像作为一类重要的矿山测量数据,其质量在很大程度上受到矿区成像环境的影响。鉴于小波变换的图像多尺度分析特性,提出了一种小波域遥感图像模糊隶属度阈值去噪算法。该算法首先对失真的矿山遥感图像采用均值滤波算法进行预处理,消除一部分图像噪声。然后对预处理后的图像进行3层小波分解,对于基本不受噪声污染的低频小波分解系数不作处理,根据高频小波分解系数的噪声分布特征,设计了模糊隶属度阈值去噪模型用于去除其中的噪声,该模型对小波软阈值去噪模型进行了2点改进:1根据图像局部区域噪声信息与非噪声信息难以有效区分的情况,设计了模糊隶属度因子,通过设定特定的小波阈值,对不同的高频小波分解系数是否含有噪声进行自适应判定;2顾及到图像小波分解层数以及各高频小波分解系数的幅值,对经典小波贝叶斯阈值计算方法进行了改进。最后将原始低频小波分解系数与去噪后的高频小波分解系数进行重构,得到高清晰度的矿山遥感图像。采用山东兖州矿区的1幅遥感图像进行试验,并引入了边缘保持指数(Edge protection index,EPI)、信噪比(Signal noise to ratio,SNR)进行去噪效果评价,结果表明:所提算法对于失真的矿山遥感图像的去噪效果明显优于小波硬阈值、软阈值去噪模型,并且相对于已有的2类改进型去噪模型,优势也较显著。
[Abstract]:As a kind of important mine survey data, the quality of mine remote sensing image is greatly affected by the imaging environment of mining area. In this paper, a wavelet domain remote sensing image fuzzy membership threshold de-noising algorithm is proposed. Firstly, the mean filter algorithm is used to preprocess the distorted mine remote sensing image. A part of the image noise is eliminated, and then the pre-processed image is decomposed by three-layer wavelet decomposition, and the low-frequency wavelet decomposition coefficient, which is basically free from noise pollution, is not processed, according to the noise distribution characteristics of the high-frequency wavelet decomposition coefficient, The fuzzy membership threshold denoising model is designed to remove the noise in the model. The wavelet soft threshold denoising model is improved by 2 points. According to the local region noise information of the image, it is difficult to distinguish the noise information from the non-noise information effectively. The fuzzy membership factor is designed. By setting a specific wavelet threshold, the wavelet decomposition coefficients of different high frequency are adaptively judged to take into account the number of wavelet decomposition layers of the image and the amplitude of each high frequency wavelet decomposition coefficient. The classical Bayesian threshold calculation method is improved. Finally, the original low-frequency wavelet decomposition coefficient and the de-noised high-frequency wavelet decomposition coefficient are reconstructed. A high definition remote sensing image of mine was obtained. A remote sensing image of Yanzhou mining area in Shandong province was used to carry out the experiment, and the edge holding index (Edge protection) was introduced to evaluate the denoising effect, and the signal-to-noise ratio (signal-to-noise ratio) was used to evaluate the de-noising effect. The results show that the proposed algorithm is better than the wavelet hard threshold and soft threshold denoising model for the distorted mine remote sensing image, and it has a significant advantage over the existing two kinds of improved de-noising model.
【作者单位】：福建工程学院交通运输学院;
【基金】：福建省中青年教师教育科研项目(编号:JA15355)
【分类号】：TD17;TP751

【相似文献】