水下激光目标的统计对消分割法

水下激光目标的识别是一个崭新的研究领域,有许多问题需要解决,其中,目标分割是关键.水下激光图像中夹杂着严重的散斑噪声,受其影响,要识别水下激光目标,就要对图像进行有效的消噪,然后进行目标分割.本文依据具有相似统计特征的噪声可抵消图像中的相应噪声这一基理,结合小波变换和统计法,提出了一种水下激光目标的统计对消分割法,以去除噪声,提取目标.实验结果表明该方法是有效可行的.

Target Segmentation from Laser Underwater Images Based on Statistic Counteraction

The recognition of laser underwater target is a new disquisitive field, in which many problems are open, and segmentation is the key problem. To recognize underwater target,segmentation must be performed efficiently because there is a lot of speckle noise in laser underwater images. According to the principle that noise with identically statistic characteristics can counteract noise in images, a new algorithm called target segmentation from laser underwater image based on statistic counteraction is proposed to remove speckle noise and extract targets with wavelet transform and statistics. Experiments demonstrate that this method is efficient and feasible.