ABSTRACT
Ancient Chinese calligraphic works have important historic and artistic values, and styles classification based on image pattern recognition is a significant issue to be addressed. To attack the drawbacks of some conventional image operators in this scenario, a method based on the fusion of GIST and Pyramid Histogram of Oriented Gradients(PHOG) feature is proposed. This method uses GIST and PHOG to extract the global features and the local contour features respectively, then feed the two features to PCA for dimensionality reduction, and connect the reduced features after PCA to form a fusion feature. Finally, the Support Vector Machine(SVM) classifier performs the style training and classification. The algorithm is verified on datasets of four tablets by calligrapher Yan Zhenqing in different periods; both the classification rates on single features and on fused features, as well as the influence of numbers of training samples on classification performance demonstrates the effectiveness of the proposed method.
- Yao C, Cheng G. Approximative Bayes optimality linear discriminant analysis for Chinese handwriting character recognition{J}. Neurocomputing, 2016, 207:346--353. Google ScholarDigital Library
- Yang W, Jin L, Xie Z, et al. Improved deep convolutional neural network for online handwritten Chinese character recognition using domain-specific knowledge{J}. 2015, 15(6):551--555. Google ScholarDigital Library
- Tao D, Jin L, Zhang S, et al. Sparse Discriminative Information Preservation for Chinese character font categorization{J}. Neurocomputing, 2014, 129(1):159--167. Google ScholarDigital Library
- Gao P, Wu J, Yuan L, et al. Fast Chinese calligraphic character recognition with large-scale data{J}. Multimedia Tools & Applications, 2015, 74(17):7221--7238. Google ScholarDigital Library
- Tao D, Lin X, Jin L, et al. Principal Component 2-D Long Short-Term Memory for Font Recognition on Single Chinese Characters.{J}. IEEE Transactions on Cybernetics, 2016, 46(3):756.Google ScholarCross Ref
- Zhang X, Nagy G. Style comparisons in calligraphy{J}. Document Recognition & Retrieval XIX, 2012, 8297(2):263--271.Google Scholar
- Zhang Y, Liu Y, He J, et al. Recognition of calligraphy style based on global feature descriptor{C}// IEEE International Conference on Multimedia and Expo. IEEE, 2013:1--6.Google Scholar
- Gao P, Wu J, Lin Y, et al. Fast Image-based Chinese Calligraphic Character Retrieval on Large Scale Data{C}// Acm/ieee-Cs Joint Conference on Digital Libraries. IEEE Press, 2014:211--218. Google ScholarDigital Library
- Imdad A, Bres S, Eglin V, et al. Writer Identification Using Steered Hermite Features and SVM{C}// International Conference on Document Analysis and Recognition. IEEE, 2017:839--843. Google ScholarDigital Library
- Vásquez J L, Travieso C M, Alonso J B. Using calligraphies features for offline writer identification{C}// International Carnahan Conference on Security Technology. IEEE, 2014:1--6.Google Scholar
- Sun R, Lian Z, Tang Y, et al. Aesthetic visual quality evaluation of Chinese handwritings{C}// International Conference on Artificial Intelligence. AAAI Press, 2015:2510--2516. Google ScholarDigital Library
- Hou S, Xu P. An algorithm of calligraphy beautification based on improved velocity and width model{C}// International Conference on Computer Science. IEEE, 2015:124--127.Google Scholar
- Liu L, Xia W, Jin L, et al. A Kai style contour beautification method for Chinese handwriting characters{C}// IEEE International Conference on Systems Man and Cybernetics. IEEE, 2010:3644--3649.Google Scholar
Index Terms
- Combination of GIST and PHOG Features for Calligraphy Styles Classification
Recommendations
Novel color Gabor-LBP-PHOG (GLP) descriptors for object and scene image classification
ICVGIP '12: Proceedings of the Eighth Indian Conference on Computer Vision, Graphics and Image ProcessingThis paper presents a novel set of color descriptors for object and scene image classification. We first introduce a new Gabor-PHOG (GPHOG) descriptor by concatenating the Pyramid of Histograms of Oriented Gradients (PHOG) of the local Gabor filtered ...
Novel Gabor-PHOG features for object and scene image classification
SSPR'12/SPR'12: Proceedings of the 2012 Joint IAPR international conference on Structural, Syntactic, and Statistical Pattern RecognitionA new Gabor-PHOG (GPHOG) descriptor is first introduced in this paper for image feature extraction by concatenating the Pyramid of Histograms of Oriented Gradients (PHOG) of all the local Gabor filtered images. Next, a comparative assessment of the ...
Learning completed discriminative local features for texture classification
We propose a novel feature representation method, namely Completed Discriminative Local Features (CDLF), for texture classification.The CDLF learn transformation matrices for texture images that maximize the mutual information between the local features ...
Comments