Research On High-quality Display Of Wide-bits Digital Images In Complex Scenes

With the development of industrial technology,in order to retain more information in the natural scene and obtain richer details,digital images with high bit width(> 8 bits)are becoming increasingly common in engineering.However,currently the commonly used display devices are still 8-bit,and it is difficult to clearly display rich information in wide-bits image.Therefore,the wide-bits image needs to be processed to adapt it to the normal display.The research purpose of this paper is to realize the dynamic range adjustment of wide-bits image,map the high dynamic range of the wide-bits image suitable for display,effectively maintain and enhance the information in the image,and avoid visual blemishes for convenient human observation or machine processing.This article mainly conducts the following research:1.Analyze and explore the characteristics and processing difficulties of wide-bits digital images in complex scenes,investigate and analyze the current status and development of high-bit-width digital image processing at home and abroad.It sorts out and analyzes the classic and effective algorithm theories in the field of image enhancement and image denoising.2.Aiming at the contradiction between the local detail retention and global consistency of the existing wide-bits image mapping algorithm,the algorithm relies too much on experience to manually select parameter values,and is not robust to different scenes.This paper proposes an adaptive dynamic range adjustment and detail enhancement method.This method uses the gradient energy function to guide the adaptive generation of a series of globally consistent images;on this basis,each of the series of images is decomposed to a base layer and multi-scale detailed layers.In order to effectively retain the image detail information and ensure the overall consistency of the image,adaptive fusion of decomposed images based on information entropy guidance is processed.The effect is verified based on public data sets and actual engineering images,through subjective analysis,information entropy,Tenengrad function and other objective evaluations;the results show that the method is effective and robust in dynamic range compression and image detail information enhancement with less parameter adjustment and strong adaptive ability.3.Aiming at the problem of image degradation caused by noise in actual scenes,a denoising method based on total variation model and multi-scale guided filter is proposed.This method uses a total variation model to effectively remove image noise and retain image information.Based on the denoised image,a series of guide images are generated,and the guide images are iteratively decomposed and total variation denoising,which effectively removes noise and avoids oversmoothed;and multi-scale information fusion realizes the integration of noise image denoising,dynamic range compression and detail preservation.The algorithm is verified by gaussian additive noise images and engineering noise images.This method effectively suppresses noise while maintaining the detailed information of complex scenes.It is evaluated by quantitative indicators such as peak signal-to-noise ratio PSNR and structural similarity SSIM.The performance is good.In summary,the research is aimed at high-quality display of wide-bits images which in large dynamic range,and an adaptive dynamic adjustment image display algorithm is proposed,which effectively and uniformly compresses the dynamic range of the mapped image while maintaining image detail information.Aiming at the inevitable noise problem in engineering projects,a large dynamic range display algorithm based on total variation denoising and guided filtering is proposed to achieve high-quality display of noise images.Experiments show that the algorithm proposed in this paper has a significant improvement in image quality and robust adaptability.