红外目标检测技术研究-机械电子工程专业论文.docxVIP

摘 摘 要 红外图像点目标的检测问题是近年来图像处理与目标识别领域的一个难点。 当红外图像传感器和目标之间距离较远时，目标在焦平面上成像面积小，且信噪 比低，目标易被噪声淹没，给检测带来了极大的难度。本文在总结了前人研究的 基础上，着重研究了小波变换在目标检测技术中的应用。利用小波的时一频特性 和多分辨率分析性能，以及背景信号、白噪声信号、点目标信号在小波各级变换 之间所呈现出的不同相关性，提出了一种基于小波区域相关滤波算法的点目标检 测方法，并结合自适应阈值方法及分割、聚类的方法，能很好地分割出目标点。 本算法克服了使用传统的滤波方法对边缘信号和噪声无法有效区分的缺陷。 考虑到工程中的应用，本文对基于DSP和FPGA的红外图像目标检测硬件系 统进行了深入的讨论，已成功完成了硬件设计任务，并在此基础上对课题的研究 前景进行展望。 关键词：点目标小波变换 多分辨率分析 区域相关 自适应阈值 ABSTRACTThe ABSTRACT The technology of point target detection in infrared image sequences is a difficult subject in the field of image processing and target recognition in recent years．Due to the long distance between infrared image sensor and the point target，it looks like a point source and occupies only one pixel in the image obtained by the sensor without any shape characteristics．The SNR of the image is very low and the target is overflowed by noise，SO the detection becomes quite difficult．Combined with many detection algorithms given by the former,the applications of wavelet transform in point target detection are mainly discussed in this paper．According to different correlation of background signal and noise between neighboring scales，the approach of point targets detection based on wavelet correlation algorithm has been tried and investigated With the time—frequent analysis and multi—resolution analysis ofwavelet transform．Using the adaptive threshold methods and the segmentation algorithms，the point target carl be separated from the background．This algorithm call overcome the shortcoming that traditional filtering methods Can not distinguish between the noise and edge signals of the infrared image． Considering the application in engineering，this paper also discusses a hardware system for point target detecting based on Field Programmable Gate Array(FPGA)and Digital Signal Processing(DSP)，and had finished the hardware board design．In the end． the prospect ofthe project is as well presented． Keywords：Point target Wavelet transform Multi-resolution analysis Area c