Efficient Discriminative K-SVD for Facial Expression Recognition

Liu, Weifeng; Song, Caifeng; Wang, Yanjiang

doi:10.1007/978-3-642-38466-0_2

Cited by 1 publication

(3 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…With regard to image classification the sparserepresentation-based classification (SRC) methods have been widely employed such as face recognition, [22][23][24] facial expression recognition, 25 weather recognition, 26 PolSAR image classification, 27 hyperspectral image classification, 28 scene image categorization, 29 satellite image classification, 30 etc. Specifically, Wright et al proposed the SRC algorithm for face recognition: training images are selected to construct a dictionary and distinguishing a new image is realized by solving the sparse coefficients corresponding to the dictionary.…”

Section: Introductionmentioning

confidence: 99%

“…For the initialization value of D, the method is similar to the classification method based on K-SVD without classifier, in which a subdictionary corresponding to each class is trained with K-SVD algorithm and subsequently the subdictionaries are concatenated to construct a whole dictionary. Then, the sparse coefficients S can be solved using the whole dictionary as initial values through K-SVD algorithm, which is used to compute the initialization values of A and W in equation (25).…”

mentioning

confidence: 99%

“…With regard to LC-KSVD1, when solving dictionary D and transformation matrix A, it adopts the same method with LC-KSVD2 except ¼ 0. After dictionary training, the sparse coefficients S can be solved so classifier W can be obtained by ridge regression method as equation (25). For test stage, the normalization and classification method are also the same with LC-KSVD2.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Machinery fault diagnosis based on time–frequency images and label consistent K-SVD

Yuan

Chen

Dong

2017

Proceedings of the Institution of Mechanical Engineers, Part C:

View full text Add to dashboard Cite

Wavelet time–frequency analysis has been widely used for machinery fault diagnosis. Mechanical vibration signals can be converted to time–frequency images using wavelet transform, so machinery fault diagnosis can be transformed to the problem of image classification. Label consistent K-SVD algorithm has been proven to be effective in image classification, which incorporates a label consistent term namely discriminative sparse code error into the objective function. Therefore, in this paper, a novel bearing fault diagnosis method based on wavelet time–frequency image and label consistent K-SVD is proposed. Firstly, continuous wavelet transform is utilized to generate wavelet time–frequency images that can fully reflect bearing fault characteristics. Then texture feature extraction based on gray level co-occurrence matrix is implemented on the wavelet time–frequency images. Finally, label consistent K-SVD is conducted for classification of the time–frequency images, and thus bearing fault diagnosis is realized. The experiment results show that the texture features based on gray level co-occurrence matrix of wavelet time–frequency images can effectively extract the fault characteristics of rolling bearings, and label consistent K-SVD performs better than other classification methods based on dictionary learning under the same parameters.

show abstract