论文部分内容阅读
激光水下目标的识别是一个崭新的研究领域 ,有许多问题亟待解决 ,其中目标提取问题是关键。由于激光照射下的水下图像含有大量的散斑噪声 ,严重影响图像的质量。依据统计规律相同的噪声可以对消图像中的噪声这一原理 ,结合小波变换和统计方法 ,提出了一种基于统计对消的水下目标分割方法 (SC)。其中 ,采用小波变换 ,对低频成分进行特殊处理 ,以增强图像中的目标信息 ;采用统计方法来充分抵消图像中的干扰噪声。实验结果表明 ,水下目标分割方法是有效且可行的。
Laser underwater target recognition is a new field of research, there are many problems to be solved, in which the problem of target extraction is the key. The underwater image under laser irradiation contains a lot of speckle noise, which seriously affects the image quality. Based on the principle that the same noise of statistics can eliminate the noise in the image, an underwater target segmentation method (SC) based on statistical cancellation is proposed combining with wavelet transform and statistical method. Among them, the wavelet transform is used to deal with the low-frequency components specially to enhance the target information in the image. The statistical method is used to fully cancel the interference noise in the image. Experimental results show that underwater target segmentation method is effective and feasible.