Open Access. Powered by Scholars. Published by Universities.®
- Institution
- Publication
- Publication Type
Articles 1 - 2 of 2
Full-Text Articles in Engineering
Color Texture Image Classification Based On Fractal Features And Extreme Learning Machine, Erkan Tanyildizi
Color Texture Image Classification Based On Fractal Features And Extreme Learning Machine, Erkan Tanyildizi
Turkish Journal of Electrical Engineering and Computer Sciences
Texture classification, especially color texture classification, is considered a significant step in segmentation and object classification. The property of color and texture is important for characterizing objects in natural scenes. Fractal dimension (FD) has many applications in the field of image compression and image segmentation. A series of FD features, such as mean, standard deviation, lacunarity, kurtosis, skewness, entropy, inverse difference moment, contrast, energy, dissimilarity, homogeneity, and maximum probability, are investigated for obtaining the maximum discrimination. In this manuscript, a methodology is proposed that is based on FD and an extreme learning machine for color texture classification. Performance of the …
Sparse Coding Based Dense Feature Representation Model For Hyperspectral Image Classification, Ender Oguslu, Guoqing Zhou, Zezhong Zheng, Khan Iftekharuddin, Jiang Li
Sparse Coding Based Dense Feature Representation Model For Hyperspectral Image Classification, Ender Oguslu, Guoqing Zhou, Zezhong Zheng, Khan Iftekharuddin, Jiang Li
Electrical & Computer Engineering Faculty Publications
We present a sparse coding based dense feature representation model (a preliminary version of the paper was presented at the SPIE Remote Sensing Conference, Dresden, Germany, 2013) for hyperspectral image (HSI) classification. The proposed method learns a new representation for each pixel in HSI through the following four steps: sub-band construction, dictionary learning, encoding, and feature selection. The new representation usually has a very high dimensionality requiring a large amount of computational resources. We applied the l1/lq regularized multiclass logistic regression technique to reduce the size of the new representation. We integrated the method with a linear …