NSCT underwater image threshold denoising algorithm based on double filtering

作者：Liu, Yue; Li, Yibing^*; Li, Ao

来源：Journal of Harbin Engineering University, 2013, 34(2): 251-255.

DOI：10.3969/j.issn.1006-7043.201203059

• Double filtering
• Image de-nosing
• Median filtering
• Non local means
• Non subsampled contourlet transform (NSCT)
• Threshold de-noising algorithms

Considering the characteristics of noise of underwater images, and the unfavorable preservation of the integrity of edge and texture information when denoising the image, a Non-Sub-sampled Contourlet Transform (NSCT) underwater image threshold denoising algorithm based on double filtering is proposed. In the algorithm, NSCT with invariant horizontal displacement was firstly applied to decompose the underwater image in multi-scale and multi-direction; then according to the noise distribution characteristics of decomposed image, by combining the frequency domain and spatial domain methods, non-local means method and median filtering algorithm were used to filter the high frequency and low frequency components, respectively; finally, a reasonable threshold was set for de-noising, the details of the image were further recovered and restructured. The experimental results show that the algorithm in this paper may apparently increase the denoising effects of underwater images.

• 出版日期2013
• 单位哈尔滨工程大学