Visibility Visible Light-far-infrared Detection Model of Main Urban Area under Thick Fog Weather Conditions in Winter
The visibility for detection of the main urban area is related to the urban planning and management,so the visibility visible light-far-infrared detection model of main urban area under thick fog weather conditions in winter is studied.The visibili-ty visible light-far-infrared original image of main urban area is decomposed by NSST to get the low frequency sub-band and high frequency sub-band of two images.The desharpening masking process method is used to eliminate the interference in low frequency sub-band.According to the regional energy fusion rule,the low frequency sub-band of visibility visible light-far-in-frared image after thick fog removal is fused.According to the Laplacian energy sum rule,the high frequency sub-band visbility visible light-far-infrared image is fused.Using the multi-scale significance test to obtain the significant region of the visibility light-far-infrared image,significant regional sample blocks are obtained through the sliding window and are input into the conv-olutional neural network with support vector machine as classifier,to get the main urban area visibility classification results.The test results show that the method can effectively remove the fog interference of visibility visible light-far-infrared images,obtain image with higher visibility,and the visibility result of the main urban area is more accurate,which can determine the construction problem of the main urban area.
thick fog in winterweather conditionmain urban areavisibilityvisible lightfar-infrared
万方数据
