Sparse representation with multi-manifold analysis for texture classification from few training images

Journal article


Sun, Xiangping, Wang, Jin, She, Mary F. H. and Kong, Lingxue. (2014). Sparse representation with multi-manifold analysis for texture classification from few training images. Image and Vision Computing. 32(11), pp. 835 - 846. https://doi.org/10.1016/j.imavis.2014.07.001
AuthorsSun, Xiangping, Wang, Jin, She, Mary F. H. and Kong, Lingxue
Abstract

Texture classification is one of the most important tasks in computer vision field and it has been extensively investigated in the last several decades. Previous texture classification methods mainly used the template matching based methods such as Support Vector Machine and k-Nearest-Neighbour for classification. Given enough training images the state-of-the-art texture classification methods could achieve very high classification accuracies on some benchmark databases. However, when the number of training images is limited, which usually happens in real-world applications because of the high cost of obtaining labelled data, the classification accuracies of those state-of-the-art methods would deteriorate due to the overfitting effect. In this paper we aim to develop a novel framework that could correctly classify textural images with only a small number of training images. By taking into account the repetition and sparsity property of textures we propose a sparse representation based multi-manifold analysis framework for texture classification from few training images. A set of new training samples are generated from each training image by a scale and spatial pyramid, and then the training samples belonging to each class are modelled by a manifold based on sparse representation. We learn a dictionary of sparse representation and a projection matrix for each class and classify the test images based on the projected reconstruction errors. The framework provides a more compact model than the template matching based texture classification methods, and mitigates the overfitting effect. Experimental results show that the proposed method could achieve reasonably high generalization capability even with as few as 3 training images, and significantly outperforms the state-of-the-art texture classification approaches on three benchmark datasets.

Keywordstexture classification; sparse representation; manifold learning; multi-manifold analysis; few training image
Year2014
JournalImage and Vision Computing
Journal citation32 (11), pp. 835 - 846
PublisherElsevier Ltd
ISSN0262-8856
Digital Object Identifier (DOI)https://doi.org/10.1016/j.imavis.2014.07.001
Scopus EID2-s2.0-84906552374
Page range835 - 846
Research GroupInstitute for Learning Sciences and Teacher Education (ILSTE)
Publisher's version
File Access Level
Controlled
Place of publicationUnited Kingdom
Permalink -

https://acuresearchbank.acu.edu.au/item/880v5/sparse-representation-with-multi-manifold-analysis-for-texture-classification-from-few-training-images

Restricted files

Publisher's version

  • 94
    total views
  • 0
    total downloads
  • 5
    views this month
  • 0
    downloads this month
These values are for the period from 19th October 2020, when this repository was created.

Export as

Related outputs

Teachers' ratings of social skills and problem behaviors as concurrent predictors of students' bullying behavior
Elliott, Stephen N., Hwang, Yoon-Suk and Wang, Jin. (2019). Teachers' ratings of social skills and problem behaviors as concurrent predictors of students' bullying behavior. Journal of Applied Developmental Psychology. 60, pp. 119 - 126. https://doi.org/10.1016/j.appdev.2018.12.005
Why choose teaching? A matter of choice : Evidence from the field
Wyatt-SmithWyatt-Smith, Claire, C., Wang, Jin, Alexander, Colette, Du Plessis, Anna, Hand, Kirstine and Colbert, Peta. (2017). Why choose teaching? A matter of choice : Evidence from the field Australia: Institute for Learning Sciences and Teacher Education, Australian Catholic University.
Probabilistic latent semantic analysis for multichannel biomedical signal clustering
Wang, Jin and She, Mary. (2016). Probabilistic latent semantic analysis for multichannel biomedical signal clustering. IEEE Signal Processing Letters. 23(12), pp. 1821 - 1824. https://doi.org/10.1109/LSP.2016.2623801
An incremental algorithm for discovering routine behaviours from smart meter data
Wang, Jin, Cardell-Oliver, Rachel and Liu, Wei. (2016). An incremental algorithm for discovering routine behaviours from smart meter data. Knowledge-Based Systems. 113, pp. 61 - 74. https://doi.org/10.1016/j.knosys.2016.09.016
Patient admission prediction using a pruned fuzzy min--max neural network with rule extraction
Wang, Jin, Lim, Chee Peng, Creighton, Douglas, Khorsavi, Abbas, Nahavandi, Saeid, Ugon, Julien, Vamplew, Peter, Stranieri, Andrew, Martin, Laura and Freischmidt, Anton. (2014). Patient admission prediction using a pruned fuzzy min--max neural network with rule extraction. Neural Computing and Applications. 26(2), pp. 277 - 289. https://doi.org/10.1007/s00521-014-1631-z
Multichannel biomedical time series clustering via hierarchical probabilistic latent semantic analysis
Wang, Jin, Sun, Xiangping, Nahavandi, Saeid, Kouzani, Abbas, Wu, Yuchuan and She, Mary. (2014). Multichannel biomedical time series clustering via hierarchical probabilistic latent semantic analysis. Computer Methods and Programs in Biomedicine. 117(2), pp. 238 - 246. https://doi.org/10.1016/j.cmpb.2014.06.014
Human identification from ECG signals via sparse representation of local segments
Wang, Jin, She, Mary, Nahavandi, Saeid and Kouzani, Abbas. (2013). Human identification from ECG signals via sparse representation of local segments. IEEE Signal Processing Letters. 20(10), pp. 937 - 940. https://doi.org/10.1109/LSP.2013.2267593
Scale invariant texture classification via sparse representation
Sun, Xiangping, Wang, Jin, She, Mary F. H. and Kong, Lingxue. (2013). Scale invariant texture classification via sparse representation. Neurocomputing. 122, pp. 338 - 348. https://doi.org/10.1016/j.neucom.2013.06.016
Biomedical time series clustering based on non-negative sparse coding and probabilistic topic model
Wang, Jin, Liu, Ping, She, Mary F. H., Nahavandi, Saeid and Kouzani, Abbas. (2013). Biomedical time series clustering based on non-negative sparse coding and probabilistic topic model. Computer Methods and Programs in Biomedicine. 111(3), pp. 629 - 641. https://doi.org/10.1016/j.cmpb.2013.05.022
Sparse representation of local spatial-temporal features with dimensionality reduction for motion recognition
Wang, Jin, Sun, Xiangping, Liu, Ping, She, Mary F. H. and Kong, Lingxue. (2013). Sparse representation of local spatial-temporal features with dimensionality reduction for motion recognition. Neurocomputing. 115, pp. 150 - 160. https://doi.org/10.1016/j.neucom.2013.01.012
Unsupervised mining of long time series based on latent topic model
Wang, Jin, Sun, Xiangping, She, Mary FH, Kouzani, Abbas and Nahavandi, Saeid. (2013). Unsupervised mining of long time series based on latent topic model. Nurocomputing. 103, pp. 93 - 103. https://doi.org/10.1016/j.neucom.2012.09.008
Supervised learning probabilistic latent semantic analysis for human motion analysis
Wang, Jin, Liu, Ping, She, Mary F. H., Kouzani, Abbas and Nahavandi, Saeid. (2013). Supervised learning probabilistic latent semantic analysis for human motion analysis. Neurocomputing. 100, pp. 134 - 143. https://doi.org/10.1016/j.neucom.2011.10.033
Bag-of-words representation for biomedical time series classification
Wang, Jin, Liu, Ping, She, Mary F. H., Nahavandi, Saeid and Kouzani, Abbas. (2013). Bag-of-words representation for biomedical time series classification. Biomedical Signal Processing and Control. 8(6), pp. 634 - 644. https://doi.org/10.1016/j.bspc.2013.06.004
Intelligent clothing for automated recognition of human physical activities in free-living environment
Wu, Yuchuan, Chen, Ronghua, Wang, Jin, Sun, Xiangping and She, Mary F. H.. (2012). Intelligent clothing for automated recognition of human physical activities in free-living environment. Journal of the Textile Institute. 103(8), pp. 806 - 816. https://doi.org/10.1080/00405000.2011.611641