| With the promotion and application of intelligent cultural tourism platform in the construction of digital society,integral imaging products have been developed rapidly in scenic spots,museums and other places.At present,there is a problem of uneven quality in the construction of virtual exhibition cases using integral imaging.Only deeper investigation on this problem can effectively control and improve the quality of integral imaging.Therefore,aiming at the problems of integral imaging product quality and panoramic image mosaic process,the integral imaging quality assurance method based on feature point mosaic is studied theoretically and experimentally.The main research contents and conclusions are as follows:(1)The construction process of integral imaging and its existing quality problems and requirements are analyzed,and the quality assurance model of integral imaging is developed.The quality improvement scheme is designed,the quality of mosaic sequence images is improved,and the process quality of integral imaging mosaic is improved.In addition,the image quality is evaluated from two aspects: one is the quality assurance of mosaic sequence images,and the other is the quality evaluation of panoramic images.(2)Aiming at the quality problems of mosaic sequence images such as low contrast and serious color distortion,uniform light processing is carried out on mosaic sequence images to equalize the gray level of the image,improve the difference of brightness and color of the image,and enhance the user’s visual experience.Simultaneously,aiming at all kinds of noise problems existing in mosaic sequence images,the mosaic sequence images are smoothed,and the neighborhood averaging method and median filtering method are compared to the noise processing effect,and finally confirm that the median filtering method is selected to remove the noise of mosaic sequence images and reduce the phenomenon of blurred images.(3)Aiming at the problem of misplacement gouging after integral imaging mosaic,an improved method for integral imaging registration and fusion was proposed.In the aspect of registration,feature point pairs are accurately registered by improving the RANSAC(Random Sampling Consensus)algorithm.In terms of fusion,the pyramid transform fusion technology is used to fuse the image,which solves the problems of integral imaging distortion and obvious mosaic traces.In addition,by comparing the image mosaic algorithm with the traditional algorithm,it is found that the improved method reduces the running time of the algorithm and improves the accuracy of the matching set.(4)The feasibility of integral imaging quality assurance model is developed and verified by screening image quality evaluation indexes.On the one hand,the objective quality evaluation of mosaic sequence images is analyzed,and the results show that the preprocessed mosaic sequence images meet the requirements of subsequent mosaic.On the other hand,subjective and objective evaluation analysis was conducted for integral imaging cases.The average score of subjective evaluation was 3.5 points,the mean value of peak signal-to-noise ratio of objective evaluation was about 38.1487 d B,and the mean value of structural similarity was about 0.9074,indicating that the integral imaging generated by the case did not have large distortion,and the registration and fusion effect was good. |
