中波1280×1024红外成像组件设计(特邀)

Design of mid-wave 1 280×1 024 infrared imaging components(Invited)

随着红外器件和成像技术的不断发展,各种夜视系统对百万像素的中波红外成像组件的需求越来越多。基于国产15μm 1 280×1 024中波HgCdTe探测器,以探测器和杜瓦自身包络为基准,突破小体积、轻量化、一体化设计,研制出了紧凑型双FPGA处理平台的百万像素中波红外成像机芯组件,组件尺寸155 mm×95 mm×95 mm,质量为1 400 g,支持SDI、Cameralink接口输出;在该平台上实现盲元替换、非均匀校正、降噪、细节增强、动态范围压缩、局部增强等实时图像处理算法,针对传统的红外成像算法提出了基于残差的非均匀校正算法与自适应局部增强算法,提升组件的成像性能。测试试验表明:组件实时输出分辨率为1 280×1 024像素的高质量低噪声的红外图像,噪声等效温差(NETD)<30 mK,组件满足高温60℃,低温-40℃工作要求,组件所采用的改进处理算法,最终输出图像提升明显。

As infrared devices and technologies develops, the demand for megapixel mid-wave infrared imaging components of various night vision systems is rising. The megapixel mid-wave infrared imaging component based on the domestic 1 280×1 024 medium wave(15 μm) infrared HgCdTe detector was developed, which took the detector and the Dewar’s own envelope as the benchmark and made a breakthrough in small form-factor and lightweight integrated design. The component with the size of 155 mm×95 mm×95 mm and the weight of 1 400 g were considered to collect and process the information from the large area of Infrared Focal Plane Arrays(IRFPA). Thus, a low-noise image acquisition and processing hardware platform based on dual FPGA architecture was brought up, which supported SDI/Cameralink interface output. In the section of algorithm, the optimized non-uniformity correction(NUC) algorithm of the focal plane pixel response and the acquisition circuit, as well as the contrast limited adaptive histogram equalization(CLAHE) method compared with traditional image enhancement technologies were proposed. In addition, blind pixel replacement, denoising,latitude reduction and further procedures were implemented to make the image quality improving. Test results show that the noise equivalent temperature difference(NETD) of the component is lower than 30 mK, and the detector performs stably at the temperature from -40 ℃ to 60 ℃. The proposed improved algorithm is fruitful in raising the quality of both thermal imaging and infrared system performance.