Analysis sparse coding models for image-based classification

Abstract

Data-driven sparse models have been shown to give superior performance for image classification tasks. Most of these works depend on learning a synthesis dictionary and the corresponding sparse code for recognition. However in recent years, an alternate analysis coding based framework (also known as co-sparse model) has been proposed for learning sparse models. In this paper, we study this framework for image classification. We demonstrate that the proposed approach is robust and efficient, while giving a comparable or better recognition performance than the traditional synthesis-based models.Index Termsanalysis sparse coding models, efficient sparse coding, image classification

Collaborative representation based face classification exploiting block weighted LBP and analysis dictionary learning

“…where Y is the matrix containing all the input images, W denotes the coefficient matrix that is initialized as a label matrix of the training samples, and T is the parameter used for constricted sparsity. To ensure that the solution is solvable and well-regularized, the set Γ is constrained to be unity row-wise norm or relatively small Frobenius norm [26].…”

“…for image representation and denoising. Shekhar [26] made improvements by adding a full-rank constraint to the analysis dictionary. Meanwhile, local feature based dictionary learning methods have been also developed to enhance the performance of collaborative representation based pattern classification.…”

Analysis Dictionary Learning: an Efficient and Discriminative Solution

“…To evaluate our proposed DCADL, we carry out a comparative study with the following methods: The first one is ADL+SVM [16], which serves as a baseline. LC-KSVD [6] is a state-of-the-art SDL.…”

Analysis Dictionary Learning Based Classification: Structure for Robustness

“…To the best of our knowledge, few attempts have been carried out for task-driven ADL. For example, in [27], Shekhar et al [27] learned an analysis dictionary and subsequently trained SVM for the digital and face recognition tasks. Their results demonstrate that ADL is more stable than SDL under noise and occlusion, and achieves a competitive performance.…”