非制冷红外FPA驱动与预处理研究

发布时间：2018-03-28 15:09

本文选题：非制冷　切入点：红外焦平面阵列　出处：《中国科学院大学(中国科学院光电技术研究所)》2017年硕士论文

【摘要】：非制冷红外FPA(Focal Plane Array,焦平面阵列)具有体积小、功耗低、价格低廉等优点,成为很多便携式红外成像系统的首要选择,已经被广泛应用到军用及民用领域。而受到制造工艺和工作原理的限制,非制冷红外FPA工作状态容易受到其工作环境的影响,且还面临红外成像中难以避免的非均匀性和图像对比度低的问题,因此必须对其进行改善。本文通过对非均匀性校正和图像增强算法的研究,以实验室研制的UL03162非制冷红外相机为基础,研究了非制冷红外FPA的驱动和预处理技术对红外成像的影响,提出对应的改进方法。本文主要完成以下工作:(1)首先介绍了非制冷红外相机的硬件系统结构,从传感器的参数,系统的存储能力和计算能力等方面验证了实现红外成像的可行性,并阐述了几个重要芯片在系统中的主要功能,为后文算法的实现及实验测试奠定了基础。(2)介绍了红外传感器的初始化过程,并通过分析红外传感器和其配置的A/D的时序,结合所设定的输出标准,为其设计了合适的初始化参数,驱动脉冲和同步信号。(3)分析出电压纹波和敏感电压参数对图像的影响,然后根据纹波设计对应的滤波电路,通过实验标定敏感电压的可用范围,获得了适用于较宽温度范围的敏感电压值。(4)针对非制冷红外焦平面的非均匀性问题,通过分析两种工程实用算法的适用环境,以输出图像的均方根噪声为评判准则,比较两点校正和一点校正的可取之处,选定可行的算法并计算相应的校正系数;针对图像对比度低的问题,总结了红外传感器所成的原始图像的直方图特点,然后比较几种主流的基于直方图的图像增强算法,提出一种简化后的直方图算法,并将算法移植到FPGA上实现。(5)最后,确定了温度波动对焦平面成像的影响,测量得到所成图像与焦平面温度之间的关系,通过实验获得了不同温度下的补偿系数,并根据焦平面温度进行实时补偿,同时为了方便测试和数据的直观,设计了测试软件,可以实现参数的实时更新和显示等。
[Abstract]:Uncooled infrared FPA(Focal Plane array (focal plane array) has many advantages, such as small size, low power consumption, low price and so on, so it has become the first choice of many portable infrared imaging systems. It has been widely used in military and civil fields. Due to the limitation of manufacturing technology and working principle, uncooled infrared FPA is vulnerable to the influence of its working environment. It is also faced with the problems of inhomogeneity and low contrast in infrared imaging, so it is necessary to improve it. In this paper, we study the algorithms of nonuniformity correction and image enhancement. Based on the UL03162 uncooled infrared camera developed in laboratory, the effects of the driving and preprocessing techniques of uncooled infrared FPA on infrared imaging are studied. This paper introduces the hardware system structure of the uncooled infrared camera and the parameters of the sensor. The storage capacity and computing ability of the system verify the feasibility of infrared imaging, and describe the main functions of several important chips in the system. The initialization process of infrared sensor is introduced. By analyzing the sequence of infrared sensor and its configuration of A / D, and combining with the output standard set, the paper introduces the initialization process of infrared sensor. A suitable initialization parameter is designed for it, driving pulse and synchronous signal. (3) the influence of voltage ripple and sensitive voltage parameter on image is analyzed. Then, according to the filter circuit corresponding to ripple design, the usable range of sensitive voltage is calibrated by experiment. In order to solve the problem of non-uniformity of uncooled infrared focal plane, the RMS noise of the output image is taken as the criterion by analyzing the applicable environment of the two practical engineering algorithms. By comparing the merits of two-point correction and one-point correction, the feasible algorithm is selected and the corresponding correction coefficients are calculated. Aiming at the problem of low image contrast, the histogram characteristics of the original image made by infrared sensor are summarized. Then compare several mainstream image enhancement algorithms based on histogram, propose a simplified histogram algorithm, and transplant the algorithm to FPGA. Finally, determine the effect of temperature fluctuation on focal plane imaging. The relationship between the image and the focal plane temperature is obtained, and the compensation coefficient at different temperature is obtained by experiments, and the real-time compensation is carried out according to the focal plane temperature. In order to facilitate the measurement and the visualization of the data, the test software is designed. Parameters can be updated and displayed in real time.
【学位授予单位】：中国科学院大学(中国科学院光电技术研究所)
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP391.41

【参考文献】