Staring array sensor model for simulation implementation

Proceedings of SPIE(2017)

引用 1|浏览0
暂无评分
摘要
A comprehensive model for staring array simulation is described. The model covers all effects from photon signal generation through to detection and processing in the staring array sensor. The model follows the signal flow from photon generation, through a staring focal plane array (FPA) from the detector, through several conversions in the read out integrated circuit (ROIC) and finally conversion to a digital signal. Spatial nonuniformity modeling for photoresponse, dark current generation and source follower offset is included. The list of noise sources includes: photon noise, quantum conversion uncertainty, dark noise, kTC noise, source follower noise and quantization noise. Several components with (simplified) nonlinear responses are also modeled: sense node capacitance variation with charge, source follower nonlinearity and nonlinearity in the digital conversion. The code implementations take images as input, applying the various processes independently on individual pixels (e.g., shot noise) or on complete images (e.g., spatial nonuniformity). Some noise sources vary temporally across frames (shot, thermal, kTC) while other noise sources are fixed across frames (fixed pattern noises). The application of the model is demonstrated by tracing the signal path from source to sensor output, with intermediate results along the path. The model is implemented in Python (as part of the pyradi open source computational radiometry module) and in a C++ image simulation. The purpose with this work is to predict what the performance of a given sensor will be in terms of image appearance, given the devices specifications and key design parameters. The execution of this work lead to the important recommendation that nonuniformity correction for infrared sensors should be performed at well fill levels corresponding to the minimum and maximum in the scene, not to fixed percentage levels in the charge well. The objective with this work is to provide a 'generic' model that can be adapted by adjusting model parameters. For more accurate modeling of specific sensors, dedicated models should be developed, but for all but the most demanding requirement, this model should be adequate in scope of detail and freedom of characteristics.
更多
查看译文
关键词
staring array performance,image simulation,photoresponse nonuniformity (PRNU),focal plane array (FPA)
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要