An Improved Dictionary Learning Super-resolution Reconstruction Method for Classroom Images
At present,the imaging of classrooms is affected by low equipment performance and complex environments,resulting in incomplete understanding of teachers and students in the teaching environment.In order to fully utilize image information and comprehensively and meticulously understand the teaching situation,this paper proposes an improved dictionary learning super-resolution reconstruction method for classroom image.By using dictionary learning algorithms to train a self constructed classroom image dataset,corresponding low rank and sparse dictionaries are obtained.The two trained dictionaries are used to reconstruct the training set images,and then participate in training to obtain residual dictionaries.Then,the three trained dictionaries are used to reconstruct low resolution images,ultimately high-resolution images are obtained.Comparative experiments are conducted between the proposed algorithm and several classic algorithms,and both visual and quantitative results show that the proposed algorithm achieved significant improvements in both PSNR and SSIM.
low rank matrix factorizationlocally linear embeddingresidual dictionaryimage super-resolution
国家科技期刊平台
NSTL
万方数据
