Research On Mid-wave Infrared Imaging Technology Based On Scene Dynamic Correction

Infrared imaging technology has important applications in the fields of photoelectric reconnaissance and search,detection guidance and tracking.At present,infrared imaging systems commonly use calibration based methods to achieve non-uniformity correction.In order to adapt to the drift and the changing scene,it is necessary to use the block to make the calibration constantly,which will lead to the transient interruption of the image.For the high-speed maneuvering target,it may cause the tracking loss,which is a very serious hidden danger.Therefore,this paper develops a the scene-based non-uniformity algorithm and proposes a non-uniformity correction algorithm based on the time-domain histogram.An infrared imaging processing circuit was designed based on the infrared focal plane of the model MARS MW K508,and the proposed method was verified.Non-uniformity correction,blind element substitution,and image enhancement are the main contents of infrared image processing.The non-uniformity correction algorithm based on the time-domain histogram describes the non-uniformity from the perspective of the probability density function of focal plane input and response,and proves that the probability density function has an equivalence relationship with the time-domain histogram,using the histogram to solve a large number The storage problem of data is a major advantage of the algorithm's hardware implementation.The blind filter is detected by the median filter algorithm,and the blind element substitution by the marker bit mode avoids the disadvantage of interrupted substitution.Combined with the histogram characteristics of infrared images,the automatic contrast enhancement method was discussed.Guided-filter high-frequency detail enhancement algorithm is used to solve the problem of detail loss in infrared image output.Adding LoG edge operator to weighted guided filter can effectively avoid the defect that the algorithm is easy to produce "halo"The infrared imaging processing circuit designed in this paper can complete the functions of detector driving,A/D signal acquisition,temperature control,real-time signal processing and so on.Differential signal transmission and two-stage filter circuit are designed to reduce the introduction of board level noise,thus reducing the noise equivalent temperature difference(NETD)of the system.The PID closed loop temperature control circuit is used to maintain the detector's temperature at 77 + 0.04K accuracy.The 4 stage pipelined structure is designed for signal processing.This structure integrates the non uniformity correction,blind element replacement,contrast enhancement and detail enhancement into one module,which greatly simplifies the processing flow and ensures the real-time output of the video signal.The hardware design and performance test of the refrigeration infrared imaging circuit designed in this paper are carried out.The test result shows that the NETD of the system is 31mK,and the imaging quality is high and the real-time performance is good.The feasibility of non-uniformity correction algorithm based on histogram in time domain is verified.