Research On Key Technologies Of Dual Band Infrared Image Fusion System

After years of efforts,the infrared imaging detection system has been developed to the third generation: dual/multi band infrared image fusion system.The core of the system is to make use of the response characteristics of the scene in different bands.The scene can have a more objective and complete description after image fusion,thus improving system performance.In this paper,the key technologies of dual band infrared image fusion system are studied.The main research contents include the following aspects:A real-time signal processing platform for dual band infrared fusion is constructed.It is mainly divided into three subsystems: High speed data transmission system,Multi core DSP real time signal processing system and Infrared video real-time display system.High speed data transmission system uses RocketIO technology to improve the bandwidth of data transmission,realizing high speed and reliable transmission of data in the system.Multi core DSP real-time signal processing system uses the architecture of multi-core DSP plus FPGA to improve the ability of real-time signal processing.Infrared video real-time display system which can display real-time high-resolution infrared images through the video interface is designed by single board.A subband decomposed multi-scale Retinex infrared image enhancement algorithm based on adaptive guided filtering is proposed.In this method,the adaptive guided filtering is combined with the subband decomposed multi-scale Retinex algorithm.The illumination component is estimated more accurately by using the adaptive guided filtering.Experiments show that the algorithm can effectively eliminate the halo and enhance the image details.An infrared image fusion algorithm based on visual saliency and NSCT is proposed.Firstly,the original images are decomposed by NSCT to obtain the multi-scale and multi direction sub-band images.Then,the saliency detection results of the images are used to guide the fusion of each subband image.Finally,the final fused image is obtained by NSCT inverse transform.The experimental results show that the algorithm can better preserve the target region information and highlight the target features.Compared with the traditional method,the fused image by our algorithm is better than the traditional one,having clearer texture details.A new algorithm for infrared dim target detection based on background suppression and image fusion is proposed.Firstly,the background suppression of each band image is carried out.Then,the effect of image registration error is reduced by using morphological dilation.Finally,the dual band images are superimposed.Experiments show that the algorithm can effectively improve the ability of target extraction.