Scene image classification with biased spatial block and pLSA

作者：Ji Zhong^*; Jing Peiguang; Wang Jing; Su Yuting

来源：International Journal of Digital Content Technology and Its Applications, 2012, 6(1): 398-404.

DOI：10.4156/jdcta.vol6.issue1.48

• Bag-of-visual words
• Bow representations
• Classification performance
• Linear coding
• Probabilistic latent semantic analysis
• Scene image classifications
• Semantic features
• Spatial constraints

Scene image classification is a fundamental problem in the fields of computer vision and image understanding. A novel scene image classification method based on biased spatial block information and an improved coding approach in bag-of-visual-words (BOW) model is proposed. The spatial constraints biased to central object regions are employed to achieve better discrimination power for image classification. Locality-constrained linear coding approach instead of traditional vector quantization is adopted. And then, we apply probabilistic Latent Semantic Analysis (pLSA) to a BOW representation for each image to obtain more compact and semantic features. Finally, scene images are classified by support vector machine with these features. Experimental results show that our method has satisfactory classification performances on a popular dataset.

• 出版日期2012
• 单位天津大学