Image Fusions Algorithm Of Infrared And Visible Light Based On PCNN

Infrared images are based on thermal radiation characteristics,the target can still be captured under low light or bad weather conditions,but the scene details and texture information performance are poor.While visible light images are based on reflection characteristics,including more details and textures Information,the resolution is higher,but in low light or blocked,the target information cannot be captured.Therefore,the fusion of infrared and visible light images is conducive to the synthesis of target information and rich scene information,and provides an effective basis for target detection and recognition in future.Infrared and visible light images fusion is widely used in target detection,security monitoring,automatic target recognition,remote sensing,it resource detection and plays an important role.Pulse Coupled Neural Network(PCNN)has the characteristics of synchronous pulse and global coupling,and is currently widely used in the field of image fusion.To solve the problems in the current infrared and visible light images fusion algorithm,two PCNN-based infrared and visible light images fusion algorithms are proposed:(1)Aiming at the problem that the traditional images fusion based on multi-scale transform has low contrast,edge details and other information retention is not ideal,a fusion algorithm based on non-downsampling Contourlet transform domain combined with adaptive fuzzy logic and adaptive PCNN is proposed.First,the non-downsampling Contourlet transform images are used for multi-scale decomposition.Then,in order to obtain the contour information in the low frequency subband,the fuzzy logic fusion rule is used.The fusion rule of adaptive PCNN is proposed in the high-frequency subband coefficients.Finally,the non-downsampling Contourlet reconstruction is performed to obtain the fusion image of infrared and visible light.The final experimental results show that the fusion algorithm can better highlight the target information of the fused image,provide rich background details,and achieve a good fusion effect in the clarity of the fused image and human vision.(2)In order to further improve the contrast and sharpness of the fused image,a algorithm based on rolling filtering and Unit-Linking PCNN is proposed.First,the infrared and visible images are decomposed to obtain the base layer and the detail layers by rolling filtering.Second,there is residual low-frequency information in the base layer,which can be used to control the overall appearance of the fused image,so a fusion rule based on regional energy rules is used in the base layer.A fusion rule of the improved adaptive Unit-Linking PCNN is adopted in the detail layers,it can retain the details and edge information of the source images well.Finally,the reconstructed images are obtained.The final experimental results show that the fusion algorithm can better highlight the target and background information of the image and improve the clarity of the fusion image.Based on the fusion algorithm,the infrared and visible light images fusion system based on PCNN is designed and implemented.The system compares the two fusion algorithms with other fusion algorithms and provides technical support for the application of infrared and visible light images fusion.