The Research Of High Dynamic Range Image Synthesizingand Tone Mapping

High dynamic range image has become a hot reaserch topic. Because of its high dynamic range, realistic visual effect and easy to capture, the high dynamic range image is very important in the field of digital image processing. high dynamic range image processing method. With high dynamic range image research technology is becoming mature,the problem of how to obtain and synthesize high dynamic range images by using ordinary camera equipment is constantly updated. However, the effect of noise on the image quality is also important in the process of synthesizing high dynamic range images. So this paper presents a method to solve the noise in the process of high dynamic range image synthesis.According to the traditional linear high dynamic range image compression technology is improved, making the tone mapping high dynamic range image is simple and effective. To solve these problems, this paper makes an explicit research in this direction. In addition, we propose two solutions to deal with these problems:1. In order to solve the noise problem that the image quality is affected by the process of the high dynamic range image synthesis, a high dynamic range image synthesis algorithm based on multi exposure image is proposed. Firstly, according to the characteristics of the photon shot noise, by the method of weighted averaging of multiple exposure images, the image mixed with the noise problem is transformed into an average value of a multi exposure image sequence group. Through image fitting method, the camera response curve is fitted,and the brightness value of the original image is obtained according to the inverse of the response curve. According to the weighted average method, a group of multi frame images are grouped to select the appropriate group size, and the average value of the current frame is modified by the different weighted coefficients. Finally, according to the response curve of the camera, the modified pixel value and the weight equation of the high dynamic range image, the pixel value of the synthesized image is obtained, which is the real value of the image.2. In order to solve the problem of dynamic range of high dynamic range image far beyond the imaging range of digital display , this paper proposes a tone mapping algorithm based on local linear transformation. Because of the structure of the local window, the relationship between the mean and the variance of the central pixel and other pixels in the window is constructed. Therefore, based on the tone mapping algorithm of local linear transformation can suppress ghosting and flare phenomenon caused by the edge image with high contrast. A local linear transformation model is proposed in this paper to solve the shortcoming of poor image compression. Firstly, according to the known image linear compression model and the high dynamic range image luminess value,the linear transformation equation of image compression is given. Secondly, different linear transformation is used in the local area of different pixel points to achieve different scaling. In the end, the luminance value of the low dynamic image is obtained by matrix transformation,and the output low dynamic range image is obtained.