Research And Design Of Editing And Enhancement Methods For JPEG Images

With the application of computers in the aspect of data processing, digital image processing technology has been rapidly developed. Image editing techniques have been widely applied in many fields. At present, there’s a lot of image editing software at home and abroad in different languages with different features. As an important technology of image processing, image enhancement algorithm is continuously evolving, hence the development of specific processing systems is required.In image processing, image editing is one of the most fundamental function. Based on VC and OPENCV, This JPEG image processing software is built and realizes common functions of image editing, such as zooming, mirroring, perspective transformation, affine transformation, text added. And it also applies common image enhance functions like image gray, histogram equalization, frequency domain filtering.Image sharpening is used to enhance the edges and outlines of objects to make the details clearer. This paper proposes an improved Laplacian sharpening algorithm to solve the problem that images will be blurred for noise is amplified after images sharpened. It first of all, to smooth the image, and then adjust sharpening coefficient, through the early late smoothing and sharpening combined to get ideal sharpening effect. This paper gives the image quality assessment parameters used for judgment after the image processing results, such as classic of MSE, PSNR and SSIM image quality parameters.The estimation of fuzzy kernel function is one of the key factors in motion blurred image restoration, namely the estimation of blur orientation and length of the degraded image. A variety of methods has been applied to the estimation of fuzzy kernel function. By studying the spectrum of motion blurred images, this paper proposes a method to compute blur orientation based on morphology, it first corrode and expand the spectrum then estimate motion blur direction based on the angle of the bright region. According to the regular pattern that black stripes periodically appear in the spectrum of blurred images, this method calculates blur length using the projection method to get the extreme points. Experimental results show the algorithm is simple and effective. It is reliable for the motion blur parameters.