红外成像系统软件架构搭建以及可重用算法模块实现

发布时间：2018-09-07 12:58

【摘要】：红外成像无论在军事领域还是民用领域都有着极为广泛的应用。现有的红外成像系统中非制冷红外探测器由于其体积小,功耗低等优势占据着主要地位,但非制冷型的红外探测器有非均匀性差,易受环境因素干扰等问题。传统的红外成像系统难以实现较为复杂的图像处理算法,导致画面质量和应用范围不尽如人意,开发的难度也较大。本文提出了新的红外成像系统架构,以此架构为基础可以方便快捷的搭建起红外成像系统,既加快了红外系统开发的速度,又能在此基础上实现更复杂的算法。本文第一章主要介绍了红外成像技术发展概况和红外成像系统的几种不同架构。第二章简要说明了新架构的三个重要组成部分:Qsys平台,Avalon总线和Nios处理器。第三章详细介绍了 Avalon-MM和Avalon-ST总线的规范和使用。在以上内容的基础上,第四章详细介绍了如何在新架构下实现红外成像系统的各个算法模块。除了通用的格式转换和PAL显示之外,本文还着重研究了提高红外成像质量的图像处理算法,并提出了改进的直方图算法和改进的图像背景补偿算法。对这些算法进行理论推导之后,实现了这些算法并以IP核的形式应用到红外成像系统新架构中。第五章描述了在一个红外系统硬件平台上实现该架构,以较小的体积和功耗,实现了非常良好的成像质量,并且整个项目的开发周期得以大大缩短。
[Abstract]:Infrared imaging has been widely used in both military and civil fields. The existing uncooled infrared detectors occupy the main position because of their small size and low power consumption. However, the uncooled infrared detectors have some problems such as poor heterogeneity and easy to be interfered by environmental factors. The traditional infrared imaging system is difficult to realize more complicated image processing algorithm, which leads to the unsatisfactory picture quality and application range, and it is also difficult to develop. In this paper, a new infrared imaging system architecture is proposed. Based on this architecture, the infrared imaging system can be constructed conveniently and quickly, which not only speeds up the development of infrared system, but also realizes more complex algorithms on this basis. The first chapter introduces the development of infrared imaging technology and several different structures of infrared imaging system. Chapter 2 briefly describes three important components of the new architecture: the Avalon bus and the Nios processor. The third chapter introduces the specification and use of Avalon-MM and Avalon-ST bus in detail. On the basis of the above, chapter 4 introduces in detail how to implement each algorithm module of infrared imaging system under the new framework. In addition to the general format conversion and PAL display, this paper also focuses on the image processing algorithm to improve the infrared imaging quality, and proposes an improved histogram algorithm and an improved image background compensation algorithm. After theoretical derivation of these algorithms, these algorithms are implemented and applied to the new architecture of infrared imaging system in the form of IP kernel. The fifth chapter describes the implementation of the architecture on an infrared system hardware platform, with a small volume and power consumption, achieved a very good imaging quality, and the entire project development cycle can be greatly shortened.
【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】：TN216

【参考文献】