基于模糊灰度变换的水下图像增强技术研究

摘要

为提高水下图像质量,分析了高斯降采样的缺点,引入了图像尺度空间的理论,采用非线性尺度空间的P-M模型,并结合高斯降采样,构造出一种新的图像降采样方法,该方法能在保留图像边缘信息的基础上,实现去噪和尺度缩放的功能.最后,针对水下图像对比度差的特点,研究了图像增强算法,对传统的模糊增强算法进行了改进,并通过实验验证了改进的增强算法比传统的模糊增强算法更适合水下管道图像的增强处理.%In order to improve the quality of underwater images, the shortcomings of Gaussian down sampling are analyzed, and on the basis of this, an image down sampling method is presented by the theory of image scale space, P-M model with nonlinear scale space as well as Gaussian down sampling. This method realizes the denoising and scaling functions. Finally, according to the characteristics of low underwater image contrast difference, the image enhancement algorithm is researched, and the traditional fuzzy enhancement algorithm is improved. The experiment verifies that the improved enhancement algorithm is superior to the traditional fuzzy enhancement algorithm, and more suitable for underwater pipeline image enhancement processing.