The Application Of Curvelet Transform In Image Dehazing Processing

With the rapid development of science and technology in current society.the Computer Vision System is now widely used in various fields such as outdoor security monitoring and capturing illegal traffic.On sunny days,these devices can capture clear images we want.However it may become paralysed under bad weathers like foggy and rainy days and so on.This thesis mainly discusses methods,based on Dichromatic Atmospheric Coloring Model.to reconstruct the image captured by the Computer Systemu under bad weather, reducing or eliminating the fog effect. This paper made a research about two methods on defog processing.The first is Vanishing Point detected by the Curvelet.The Curvelet Transform extracts the image edge texture accurately and calculates the vanishing point from the intersection of straight line edges effectivly.The taditional algorithm on calculating the depth of focus has been improved.The new algorithm calculates every pixel's depth of focus to reconstruct the image.It solves the problem of foggy image's color distortion that is caused by using same depth of focus to reconstruct pictures. It makes the processed image look more natural.The second method of image defog processing is based on the Curvelet Transform and Retinex algorithm.The Curvelet Transform decomposes the image into some different frequency components.The high frequency component is denoised of threshold method to improve the detail coefficients and clarity.The low frequency component is improved by increasing contrast and decreasing low frequency approximate coefficient.The Curvelet Transform is used in both ways but different ways are combined with different image processing auxiliary method. The emphasis is different as well.The first method takes into account the distortion phenomenon. Thus, the processed pictures look more natural.And the second method is conducted from the human visual perception to achieve the effect of defog so that the the clarity is higher after processing.It is proved by a number of simulation experiments that the two methods has the obvious effect on the image defog processing.