基于粒子群最小二乘支持向量机的前视声呐目标识别

Forward-looking sonar target recognition based on particle swarm and least squares support vector machine

随着声成像技术的日益发展和广泛应用,利用图像声呐进行水下目标识别逐渐成为水声探测领域的重要研究方向之一。根据前视声呐图像的特性,提出了一种水下目标识别的方法。对声呐图像进行去噪和增强处理并分割图像,来获取目标所在区域、提取目标的区域形状特征;利用粒子群算法优化最小二乘支持向量机的正则化参数和核参数,构造出高性能的多分类器;输入待识别目标的特征实现分类。实验表明:优化后的最小二乘支持向量机能够准确、有效地识别出水下目标,并且具有较高的精度。

With the increasing development and wide application of acoustic imaging technology, the use of image sonar for underwater target recognition has become an important research direction in the field of underwater acoustic detection. According to the characteristics of forward-looking sonar images, a method of underwater target recognition is proposed and described as follows： Denoising and enhancing the sonar image and then segmenting it to obtain the area of the target; Extracting the regional shape features of the target; Using the particle swarm optimization （PSO） algorithm to optimize the regularization parameters and kernel parameters of the least squares support vector machine （LSSVM） and then to make a high-performance multi-classifier; Entering the characteristics of the target to be identified to achieve classification. Experiments show that the optimized least squares support vector machine can effectively identify the underwater targets with high accuracy.