Research On Shape Retrieval Algorithm Based On Regularization Framework Of Diffusion Process

With the rapid development of photographic technology and the rapid iteration of equipment,image,as a medium containing a lot of information,plays an increasingly important role in daily life.A large number of image and video data are applied in many fields,such as medical diagnosis,digital economy,industrial automation,new infrastructure and so on.Image retrieval refers to retrieving the image related to text query or visual query from the large-scale image database.According to this process,the information of the query image can be captured,and then the image dataset could be effectively managed.Therefore,it has become a significant and urgent task to retrieve the image related to the query quickly and accurately from a large number of digital images.In general,the visual retrieval problem in computer vision is directly solved by ranking pairwise(dis-)similarities,which can be obtained by comparing one or several certain features of different targets.The main point of this idea is that the more similar two objects are,the smaller distance is measured.However,as demonstrated in recent studies,for such a basic retrieval approach,the intrinsic relationship between different objects has not yet been taken into account,and it also means that the underlying manifold structure is completely ignored.As the object retrieval problem cannot be well solved by pairwise distances,many algorithms have been developed for improving visual retrieval result.In recent studies,it is demonstrated that the contextual(dis-)similarities based on diffusion process can be obtained directly by solving an optimization problem with a general regularization framework,which contains a smoothness constraint and a fitting constraint.To improve the effectiveness of visual retrieval,this paper introduces a novel smoothness constraint named triplet-based smoothness constraint into the regularization framework of diffusion process.According to the random walk model,the proposed model can be used to simultaneously regularize three elements,and provides a quite different form of high-order information.Then we proposed the triplet-based regularized diffusion process by introducing the hybrid fitting constraint into our method.The experimental results on different shape databases demonstrate that retrieval results can be effectively improved by using the proposed methods.In order to verify the effectiveness and reliability of the scheme,the experiment introduces some well-known datasets to test the proposed algorithm,and then the experimental results are further analyzed in detail.It is demonstrated that the scheme proposed in this paper can improve the efficiency of image retrieval and achieve the expected research purpose.