| The application of infrared imaging systems in military and civilian fields is becoming increasingly widespread.The operating bands of these systems contain special information that is not observable in the visible light spectrum.Infrared imaging systems can capture such information and convert it into electrical signals.By using an infrared imaging system to observe the external environment,special analysis and observations of scenes can be performed.The main control chip is the core component of an infrared imaging system.With the rapid development of semiconductor technology,new architectures of embedded chips are constantly emerging.Integrated circuits are continuously evolving towards lower power consumption,higher integration,and stronger computational capabilities.Traditional design approaches for the driving circuits of infrared detectors involve either a single signal processing chip or a discrete dual-chip solution.Both of these approaches have disadvantages such as algorithm portability issues,difficulty in achieving serial processing,insufficient communication bandwidth,and overly complex hardware circuits.Furthermore,with the continuous advancement of technology in various industries,the imaging quality requirements for infrared imaging systems have also increased.Common infrared imaging systems on the market have an baffle structure.During the imaging process,the baffle intermittently blocks the detector to calibrate the image,resulting in a brief period of blindness.This can have a significant impact in fields that require high realtime performance,such as military applications and automotive driving.Based on these two issues,this paper presents a baffle-free long-wave infrared imaging system based on Zynq-7000.The specific work is as follows:(1)Designed and developed the hardware circuit of uncooled infrared imaging system.The system requires a main control chip to perform image processing and output infrared images that are observable to the naked eye.Through research on the imaging principles of the system,the following components have been developed: a driver circuit for non-cooled infrared detectors,a signal processing circuit based on Zynq-7000 as the core,and an infrared video signal output circuit.The proposed solution incorporates an FPGA with an embedded ARM hard core,which helps reduce the circuit complexity compared to traditional infrared imaging circuits that use multiple processors.(2)Implemented were a series of hardware solutions for pre-processing algorithms in an infrared imaging system.The image information output by the infrared detector contains a significant amount of noise and has low contrast.The paper develops hardware circuits for the infrared imaging system and conducts research on denoising algorithms and enhancement algorithms based on the characteristics of infrared images.The analysis of the algorithm’s resource consumption is carried out,and ultimately,on Zynq-7000,OOC correction,traditional single-point calibration,image denoising,and image enhancement are implemented.(3)Designed was a new type of non-uniform calibration algorithm without baffle.First,the characteristics of infrared imaging process were studied.The analysis and comparison of traditional non-uniformity calibration methods and the impact of removing the baffle structure on the imaging quality of infrared images were conducted.This thsis proposes a baffle-free non-uniformity calibration algorithm based on ambient temperature.Images were captured under different ambient temperatures to study the imaging characteristics of the infrared detector.An imaging model of the infrared imaging system was established,and real-time calibration of the data was performed using this model.Experimental results demonstrate that the designed data transmission method in the proposed approach has sufficient bandwidth and is compatible with the infrared detector used in the system.The imaging effect of the system designed using this approach is excellent,resulting in high-quality infrared videos without the blind-view effect. |
