支持向量机在红外成像自动目标识别中的应用

The Application of the Support Vector Machine in Infared Imaging Automatic Target Recognition

针对实战中红外成像制导导弹面临的自动目标识别问题,引入支持向量机作为分类器,提出了一种利用二维图像识别三维目标的自动目标识别方法。仿真实验结果表明:该方法能够成功地识别三维空间中任意角度的目标,很好地解决了许多二维识别算法难以解决的三维目标识别问题。并且,该方法比传统识别方法拥有更高的识别率。

Focusing on the problem of the automatic target recognition meeted by the infared missies in the war, a automatic target recognition method, using many 2-D images to recognize the 3-D target, is proposed. And the Support Vector Machine is used as the classification. The simulation experiment results demonstrate, the method could recognize any angle target in the 3-D space successfully, and solve the 3-D target recognition problem that many 2-D recognition algorithms could not resolve. And the method has a higher recognition ratio than other traditional recognition methods.