基于Cyclone Ⅳ的红外与可见光图像融合系统的设计与研究

发布时间：2018-05-21 12:40

本文选题：红外成像 + FPGA　；参考：《南京理工大学》2017年硕士论文

【摘要】：红外与可见光图像融合技术能够提取不同波段图像的特征,有效地提高了目标探测的空间范围和探测概率,增强人们在探测或者检测领域对图像的认知和理解能力。因此,在军用和民用领域该技术都受到了广泛关注。本文以研制红外与可见光的图像融合系统为核心,对红外与可见光成像原理,图像融合算法,融合系统设计展开了相关研究。在图像的预处理方面,本文提出了一种混合域算法,满足了对运动目标的观察要求的同时,提高了降噪效果,增强了目标的边缘细节。在图像融合算法方面,考虑到人眼视觉观察的特点,本文提出了一种基于人眼视觉显著性的多尺度融合算法,本方法对于金字塔分解得到的高频和低频图像分别采用不同的融合算法进行融合计算,然后经过逆变换得到重构图像。算法的核心是利用人眼视觉特性,突出源图像中人眼视觉显著性较强的区域。通过与多种融合算法进行对比,证明了本文提出算法的有效性。本文基于融合系统的功能需求,设计了基于红外与可见光实时图像融合系统方案,完成了芯片选型,以及融合电路原理图的制作。本文对于系统电路的主要功能模块,包括,模拟视频解码芯片电路、模拟/数字视频编码模块、存储电路模块、FPGA外围配置模块以及电源设计模块做了详细的介绍。接着在原理图基础上完成了 PCB的布局布线,设计出系统PCB版图。并设计图像融合系统的软件模块划分,主要分为视频编解码模块、数据存储模块、图像处理模块。最后,完成了图像融合系统的实验分析,验证了系统的融合效果。
[Abstract]:Infrared and visible image fusion technology can extract the features of different bands of images, effectively improve the spatial range and detection probability of target detection, and enhance the ability of recognition and understanding of images in the field of detection or detection. Therefore, in the military and civil fields, the technology has been widely concerned. In this paper, the image fusion system of infrared and visible light is developed as the core. The principle of infrared and visible light imaging, the image fusion algorithm and the design of fusion system are studied. In the aspect of image preprocessing, a hybrid domain algorithm is proposed, which not only meets the requirements of moving target observation, but also improves the noise reduction effect and enhances the edge detail of the target. In the aspect of image fusion algorithm, considering the characteristics of human visual observation, a multi-scale fusion algorithm based on the salience of human vision is proposed in this paper. In this method, different fusion algorithms are used to calculate the high frequency and low frequency images obtained from pyramid decomposition, and then the reconstructed images are obtained by inverse transformation. The core of the algorithm is to make use of the human visual characteristics to highlight the region with strong visual significance in the source image. The effectiveness of the proposed algorithm is proved by comparison with many fusion algorithms. Based on the functional requirements of the fusion system, this paper designs a real-time image fusion system based on infrared and visible light, completes the chip selection, and makes the schematic diagram of the fusion circuit. In this paper, the main function modules of the system circuit, including analog video decoding chip circuit, analog / digital video coding module, memory circuit module and FPGA peripheral configuration module, as well as power design module, are introduced in detail. Then, the layout and wiring of PCB are completed on the basis of schematic diagram, and the PCB layout of the system is designed. The software module of image fusion system is divided into video codec module, data storage module and image processing module. Finally, the experimental analysis of the image fusion system is completed, and the fusion effect of the system is verified.
【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】：TP391.41

【参考文献】