弹箭电视制导图像处理技术研究

发布时间：2018-03-22 06:30

本文选题：弹箭制导　切入点：DSP　出处：《南京理工大学》2017年硕士论文　论文类型：学位论文

【摘要】：随着精确制导弹箭的不断发展和广泛使用,运动目标的实时检测与跟踪已经成为现代战争中尤为重要的一环。而其真实性与实时性的好坏决定着战争的成败。在此背景下,对弹载电视导引头电视制导技术及其核心目标检测与跟踪系统展开研究具有着重要的现实意义。因此,本文设计了一套基于DSP的目标检测与跟踪系统。首先,设计了目标检测与跟踪系统的硬件结构,主要包括DM642图像处理平台和F2812转动控制云台两部分。对各个模块进行设计、配置、调试来实现DM642对视频图像的运动目标检测与跟踪,DM642通过串口通信协议与F2812进行数据传输,F2812通过输出PWM波的形式实现对转动控制云台实时控制来驱动摄像头实时跟踪运动目标。其次,弹载电视导引头作为光电设备,必然会由于光电转换、外界环境、空间电磁场等原因使采集到的视频图像产生大量的噪声。本文根据各种图像噪声产生的原理,深入研究了均值滤波、中值滤波等几种常用滤波方法并采用非线性滤波法中的中值滤波对视频图像进行预处理,以期达到理想的效果。再次,研究了目标检测与跟踪过程。分析讨论了相邻帧差法、背景差分法、均值漂移法等多种主流的运动目标检测与跟踪算法原理及实现方法。在目标检测算法中本文采用改进的对称差分法与背景差分法相结合的方法实现运动目标与背景的分离并为后续的目标跟踪提供初始搜索框;在目标跟踪算法中本文采用均值漂移算法基础上优化的连续自适应均值漂移算法实现运动目标的实时跟踪并确定了目标在图像中位置与运动参数。最后,在所设计的目标检测与跟踪系统上进行了半实物仿真实验,实验结果表明,采用本文设计的系统能够较好的实现算法对运动目标的实时跟踪,系统具有较好的实时性与鲁棒性。为进一步研究弹载电视导引头技术奠定了基础。
[Abstract]:With the continuous development and wide use of precision guided projectiles, real-time detection and tracking of moving targets has become a particularly important part of modern warfare, and its authenticity and real-time quality determine the success or failure of the war. It is of great practical significance to study the TV guidance technology and its core target detection and tracking system of missile borne TV seeker. Therefore, a set of target detection and tracking system based on DSP is designed in this paper. The hardware structure of the target detection and tracking system is designed, which includes two parts: DM642 image processing platform and F2812 rotation control cloud head. Each module is designed and configured. Debugging to realize moving Target Detection and tracking of Video Image by DM642 DM642 transmits data with F2812 through Serial Communication Protocol F2812 realizes Real-time Control of Motion Control Cloud head by output PWM Wave, drives camera Real time with F2812. Track the moving target. Second, As an optoelectronic device, the projectile TV seeker will produce a lot of noise due to the photoelectric conversion, external environment, space electromagnetic field and other reasons. Several common filtering methods, such as mean filter, median filter and so on, are studied in depth, and the median filter in nonlinear filtering method is used to preprocess the video image in order to achieve the ideal effect. The process of target detection and tracking is studied, and the adjacent frame difference method and background differential method are analyzed and discussed. The principle and implementation method of moving target detection and tracking algorithm based on mean shift method and other mainstream methods. In this paper, the improved symmetrical difference method and background differential method are used to realize moving target and back. Separating the scene and providing the initial search box for the subsequent target tracking; In the target tracking algorithm, the continuous adaptive mean shift algorithm based on the mean shift algorithm is adopted to realize the real-time tracking of moving targets and to determine the position and motion parameters of the target in the image. The hardware-in-the-loop simulation experiments are carried out on the designed target detection and tracking system. The experimental results show that the system designed in this paper can achieve the real-time tracking algorithm of moving targets. The system has good real-time and robustness, which lays a foundation for the further study of missile-borne TV seeker technology.
【学位授予单位】：南京理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TJ765;TP391.41

【相似文献】