Research On The Algorithm Of Oblique Aerial Camera Image Defogging

Oblique aerial camera has been more widely used in the military field,cause the advantages,such as:high-altitude remote tilt detection,surveillance,coverage.In the Oblique aerial camera imaging process,the aerial image quality will degradefor the reason of atmosphere influence.The aerial image quality will degrade when the distance of the light in the aerosol is longer or the visibility is lower in the atmosphere,so the oblique camera requires defogging to remove atmospheric Image quality.Firstly,the mechanism of image fog is analyzed based on the atmospheric scattering model,and the image degradation model is obtained,which is the theoretical basis of defogging.Secondly,several kinds of defogging algorithms are compared and analyzed,and the main dark channel prior is chosen as the research direction,and the deficiency of the dark channel prior is put forward.Thirdly,the paper proposes three improvements for the deficiency of the algorithm.First,this paper divides input fog images into four types by calculating the image contrast,signal-to-noise ratio and mean.The relationship between the airlight value and the three parameters is obtained by the experimental statistic rule,and an adaptive atmospheric dark channel prior method.Second,aiming at the non-uniformity of the reconstructed image by using the traditional dark channel prior algorithm,the bilateral filtering algorithm is used to optimize the atmospheric transmittance,which ensures the restored image has good uniformity and preserves the edge detail information of the image.Third,the use of downsampling shorten the running time of the algorithm,on the one hand to solve the darkchannelmaptime too large problem,on the other hand to solve the first two improved methods to join the algorithm leading to running time longer.Finally,in order to verify the advantages and disadvantages of the improved algorithm,using image quality evaluation methods to illustrate.The subjective evaluation shows that the quality of the restored image is improved.The objective evaluation index of the improved algorithm has improved nearly three times,such as:contrast,signal to noise ratio,information entropy.From the three objective evaluation indexes of contrast,signal to noise ratio and information entropy,we can see that the restored image is nearly 4 times of the original image.Through a large number of image testing,improved algorithm running time is about the original 1/3.The improved algorithm can improve the real-time performance of the algorithm and effectively remove the influence of the atmosphere on the image quality,and uncover the veil of the image fog and obtain a better restoration image.So the algorithm will be in the military reconnaissance,geographic information mapping and other fields to play a better effect.