摘要
提出了一种基于最优阈值分割算法的水下目标自动实时识别系统。该系统首先运用去噪、图像均衡等方法对实时摄取的水下图像进行预处理。然后运用基于遗传算法优化的 Otsu(即大津方法)最优阈值分割算法对所得图像进行区域分割并提取图像的特征向量。最后采用 BP 神经网络对提取的特征向量进行自动分类从而最终确定了水下目标的类型。水槽仿真试验表明该方法能够在恶劣的环境下自动地检测水下目标,而且该方法具有较强的抗光线干扰能力和较高的准确度。
An underwater objects real-time recognition system was presented in this paper which is based on the expanded Otsu optimal thresholding segmentation algorithm.In preprocessing,denoising,image enhancement techniques were used to improve the image quality.After preprocessing,images were segmented with a region segmentation algorithm:Ostu optimal thresholding algorithm based on GA algorithm.After the feature vector being extracted from the segmented image,BP neural network was employed in classifying the underwater objects(mines,torpedoes etc).This system can automatically detect and recognize underwater objects.The effectiveness and robustness were testified by the experiments in flume.
出处
《系统仿真技术》
2005年第1期33-39,共7页
System Simulation Technology
