基于水下图像小波变换的图像阈值去噪方法的研究

Study on Image Threshold De-noising Based on Underwater Image Wavelet Transform

水下图像污染过程和机理十分复杂,不同的水下环境噪声也不尽相同。在此尝试用小波阈值去噪方法对水下图像进行去噪,力求改善图像质量。小波阈值去噪是信号处理中一种重要的去噪方法,针对常用硬阈值函数不连续的特点以及软阈值函数存在偏差的问题,提出了一种新的阈值处理方法。在Matlab中的仿真试验结果表明,新阈值方法的去噪效果无论在视觉效果上,还是在信噪比上都优于传统的硬阈值和软阈值,充分体现出小波阈值去噪方法的优越性。

The pollution process and mechanism of underwater images is very complex.The noise under different underwater environment is not the same.The wavelet threshold de-noising method is used to denoise from underwater images and to improve the image quality.The wavelet threshold de-noising is an important de-noising method in the signal processing.A new threshold processing methom based on the wavelet threshold de-noising is presented.The method can overcome the shortcomings of the hard threshold with discontinuous functions and solve the problem of the bias in soft threshold functions.The de-noising effect of the new threshold method is better than the traditional hard threshold or soft threshold method in visual effect.The results of simulation with Matlab show that the new threshold function is effective,and can embody the advantages of the wavelet threshold de-noising method more fully.