基于特征的江面轮船识别算法*

本文提出了一种江面轮船目标的特征识别方法。首先对江面轮船图像进行预处理，然后通过二维小波变换，提取出边缘轮廓，将目标物体与背景分离开来。结合本文提出的FE（Feature Extract）算法提取出轮船图像的四个特征，根据这些特征建造一个知识库，通过选取适当的知识，采用产生式规则对目标物体进行判别，排除干扰目标，从而识别出轮船目标。最后本文从图片库中抽取几张图片进行实验，相比于之前的单特征方法和Adboost方法可以看出，该方法在识别率上要高于单特征方法，在识别速度上要快于Adboost方法。

Ship targets recognition algorithm based on features

In this paper, a feature method for ship recognition was proposed. First, preprocess the image of the boats, and then put up two-dimensional wavelet transform on the image, abstract the marginal profile and separate the target from the background. Combined with FE(Feature Extract) algorithm proposed in this paper, get four features of the boats, according to the features, build a knowledge base, then choose the correct knowledge and use set of generative rules to distinguish the objects, eliminate the interfering objects, then identify the ship target. Finally, from the experiment carried on some extracted pictures, and compared with the former single feature mothed and Adboost method, It shows that this method can recognize ships more efficiently than single feature method and more effectively than Adboost method.