Design of fusion system based on infrared moving target segmentation
-
摘要: 为了提高夜视系统的质量以及目标探测性能, 设计了红外视频运动目标与可见光融合夜视侦察系统。系统在硬件处理平台上实现了基于人眼视觉的红外运动目标分割算法和基于目标特性的加权融合算法, 采用红外运动目标分割电路和融合处理电路互联的结构, 首先利用FPGA为核心的红外目标分割电路提取红外视频中的运动目标, 然后将只有红外运动目标的视频输入后端DSP融合处理电路中进行融合处理, 最后从系统中输出一系列目标形态突出、背景细节清晰的融合序列。实验结果表明: 该系统可以提高夜视融合系统目标探测性能和探测概率, 融合结果的各项评价指标提高了90%以上, 有的评价指标甚至提高了7倍以上, 可以很大程度地降低系统的虚警率。Abstract: In order to improve the fusion image quality and target detection performance of infrared and visible image fusion system, an image fusion reconnaissance system was designed which used moving target of infrared video and visual to moniter the scene. This system realized the infrared image segmentation algorithm based on human visual characteristics and weighted fusion algorithm based on target's characteristics in the hardware platform. Also, the system contained the circuit of infrared moving object segmentation and fusion processing circuit. Firstly, image segmentation circuit to extract the moving target of infrared video was designed by FPGA. Secondly, the image which only contained moving target was input to fusion processing circuit consisting of DSP. Last, a series of fusion images which had the characteristic of target form prominent and background details clear were furnished. Through the experiment, this system can both improve the target detection performance and processing speed. The image fusion evaluation increased more than 90%, and some indexes increased more than 7 times.
-
Key words:
- image fusion /
- moving object segmentation /
- fusion system /
- infrared video /
- visible light image
-
[1] [2] Liu Kun, Guo Lei, Li Huihui, et al. Fusion of infrared and visible light images based on region segmentation [J]. Chinese Journal of Aeronautics, 2009, 22: 75-80. [3] Niu Yifeng, Xu Shengtao, Wu Lizhen, et al. Airborne infrared and visible image fusion for target perception based on target region segmentation and discrete wavelet transform[J]. Mathematical Problems in Engineering, 2012,ID275138. [4] [5] Li Zhenhua, Jing Zhongliang, Sun Shaoyuan, et al. Infrared and visible dynamic image based on moving target detection[J]. Journal of Shanghai Jiaotong University, 2005, 39(8): 1304-1311. (in Chinese) 李振华, 敬忠良, 孙韶媛, 等. 基于目标检测的红外和可见光动态图像融合[J]. 上海交通大学学报, 2005, 39(8): 1304-1311. [6] [7] Zhan Qihai, Fu Rongguo, Chang Benkang. Optical axes boresight instrument for multispectral optical system[J]. Journal of Applied Opticals, 2005, 26(5): 4-6. (in Chinese) 詹启海, 富容国, 常本康. 多光谱光学系统光轴平行性组合测试装置[J], 应用光学, 2005, 26(5): 4-6. [8] [9] [10] Sun Bin, Min Chaobo, Zhang Junju, et al. A method for moving objects segmentation based on human vision perception in infrared video [J]. Optical Review, 2014, 21(1): 27-34. [11] [12] Gonzalez R C, Woods R E. Digital Image Processing[M]. 2nd ed. Beijing: Publishing House of Electronics Industry, 2003. [13] Yang Wenbin, Zhang Junju, Xu Hui, et al. Study of infrared and LLL image fusion algorithm based on the target characteristics[J]. Laser Infrared, 2014, 44(1): 56-60. (in Chinese) 杨文彬, 张俊举, 许辉, 等. 基于目标特性分析的红外与微光图像融合算法[J]. 激光与红外, 2014, 44(1): 56-60. [14] [15] Yuan Yihui. Fusion and evaluation technology of infrared and visible images[D]. Nanjing: Nanjing University of Science and Technology, 2012. (in Chinese) 袁轶慧. 红外与可见光图像融合及评价技术研究[D]. 南京: 南京理工大学, 2012. [16] [17] [18] Liu Lei, Chang Benkang. Spectral matching factors between super S25 and new S25 photo caathodes and reflective radiation of objects[J]. Applied Optics, 2004, 43(3): 616-619. [19] Chen Xiuhong, Wei Heli, Xu Qingshan. Infrared atmospheric transmittance calculation model[J]. Infrared and Laser Engineering, 2011, 40(5): 811-816. (in Chinese) 陈秀红, 魏合理, 徐青山. 红外大气透过率的计算模式[J]. 红外与激光工程, 2011, 40(5): 811-816. [20] [21] Bailey H H. Target Detection Through Visual Recognition and a Quantitative Model[M]. New York: The Rand Corporation, 1970.
计量
- 文章访问数: 238
- HTML全文浏览量: 10
- PDF下载量: 292
- 被引次数: 0