Predictive high-level feature representation based on dictionary learning

• Published in: Expert systems with applications 2017-03, Vol.69, p.101-109
• Main Authors: Phaisangittisagul, E., Thainimit, S., Chen, W. K.
• Format: Article
• Language: English
• Subjects: Algorithms; Benchmarks; Classification; Coding; Coefficients; Dictionary learning; Experiments; High-level feature representation; Image classification; K-SVD; Level (quantity); Matrix algebra; Matrix methods; Optimization; Predictions; Singular value decomposition; Sparse coding; Supervised learning; Visual tasks
• ISSN: 0957-4174; 1873-6793
• DOI: 10.1016/j.eswa.2016.10.019

•New constraint to enforce label consistency of sparse coding is proposed.•An efficient linear model of sparse coding prediction with low cost is introduced.•This method provides promising classification accuracy compared with others.•The proposed method clearly outperforms in terms of computation time.•This method can be considered to apply in real-time classification problems. A much improved computational performance of visual recognition tasks can be achieved by representing raw input data (low-level) with high-level feature representation. In order to generate the high-level representation, a sparse coding is widely used. However, a major problem in traditional sparse coding is computational performance due to an ℓ0/ℓ1 optimization. Often, this process takes significant amount of time to find the corresponding coding coefficients. This paper proposed a new method to create a discriminative sparse coding that is more efficient to compute the coding coefficients with minimum computational effort. More specifically, a linear model of sparse coding prediction was introduced to estimate the coding coefficients. This is accomplished by computing the matrix-vector product. We named this proposed method as predictive sparse coding K-SVD algorithm (PSC–KSVD). The experimental results demonstrated that PSC–KSVD achieved promising classification results on well-known benchmark image databases. Furthermore, it outperformed the currently approaches in terms of computational time. Consequently, PSC–KDVD can be considered as a suitable method to apply in real-time classification problems especially with large databases.