强噪声背景下莫尔斯信号的自动检测与识别  

Automatic detection and recognition of Morse signal in strong noise environment

作　　者：张汝波 何立刚[1] 李雪耀[1] 

机构地区：[1]哈尔滨工程大学计算机科学与技术学院,黑龙江哈尔滨150001

出　　处：《哈尔滨工程大学学报》2006年第1期112-117,共6页Journal of Harbin Engineering University

基　　金：国家自然科学基金资助项目(60475016);国防基础研究基金资助项目(K1503060414)

摘　　要：由于当前对莫尔斯报的接收主要还是采用人工方式,易受人的心理、生理因素的影响.为了减轻人的劳动强度,研究能够部分取代人的自动检测与识别系统就十分必要.基于短时傅里叶变换形成的三维谱图,采用数字图像处理的方法提取信号的特征.首先对图像进行增强以提高高灰度区和区域边界的清晰度,运用自适应阈值分割技术对图像进行二值化及数学形态学处理;然后提取图像中的区域形状特征,获得比较纯净的莫尔斯信号图像,在此基础上采用识别译码和纠错相结合的方法提高自动检测的准确性.实验表明,将一维莫尔斯信号变换到二维空间运用图像处理技术进行信号的分析是切实可行的,该算法具有更高的识别率,具有一定的实用价值.The reception of Morse telegraph signals primarily requires manual work. Thus, its accuracy is easily affected by human physiology. To improve the system and provide a partial substitute for human effort, an automatic detection and recognition system was investigated. Digital image technology was adopted to extract the characteristics of the Morse signal based on a three-dimensional spectrum formed by a short time Fourier transform. First, the image was enhanced in order to raise definition of the high grey level district and the goal edge. Adaptive threshold segmentation was used to obtain a binary image and mathematics morphology was also used to deal with the image. Then, the regional shape characteristics were extracted and purer Morse signals were obtained. On this basis, recognition and correcting decoding were combined to improve the accuracy of automatic detection. Experiment results show that it is possible to transform a one-dimensional Morse signal to two-dimensional space and then analyze it with image processing technology. The algorithm has a higher recognition ratio and offers practical capabilities.

关 键 词：莫尔斯信号 短时傅里叶变换 三维谱图 

分 类 号：TP274.2[自动化与计算机技术—检测技术与自动化装置]