摘要
针对传统雷达HRRP自动目标识别算法中识别率较低和库外样本拒判效果不理想的问题,提出了一种基于数据预处理和子像空间的新识别方法。该方法一方面通过对数据的预处理来消除噪声的影响,另一方面通过建立子像空间进一步提高了库内样本的识别率,并基于假设检验理论引入了拒判门限,用于对库外样本进行拒判。实验表明,该方法不仅在识别率方面优于传统识别算法,且能有效对库外样本进行拒判。
The traditional radar HRRP automatic target recognition algorithm to identify the sample rate is low and rejection of the outside samples is not ideal. This paper propose a new method to identify data preproeessing and child like space-based. On one hand, the method of data preprocessing eliminate the effect of noise, on the other hand, through the establishment of sub-like space, the recognition rate of the inside library of samples is improved, and the introduction of threshold based on hypothesis testing theory improves the rejection performance. Experiments show that this method is not only superior to the traditional algorithm in the recognition rate, and can effectively resist the library outside the sample sentence.
出处
《电子科技》
2014年第12期150-154,共5页
Electronic Science and Technology
