Static Image Classification based on ScSPM and LBP histogram Fourier (LBP-HF) Features

In the recent digital age, support vector machines (SVMs) that use a spatial pyramid matching (SPM) kernel have been around the globe for image classification. Although this is popular, there exists many problems in its use; for examples, nonlinear SVMs have a high complexity in training and testing. Applying the algorithms to big datasets, which holds many images, greater than a thousand is a detailed and complicated task. This paper develops an extension of the linear SPM that uses linear kernel on spatial-pyramid pooling of SIFT sparse codes (ScSPM), by generalizing vector quantization to sparse coding followed by multi-scale spatial max pooling, we propose a linear SPM kernel based on LBP-HF sparse codes. This highly-innovative approach remarkably reduces complexity of the training and testing. Our image categorization experiment results regarding classification accuracy, suggests that a linear SPM, based on a sparse coding of the LBP-HF descriptors, significantly outperforms the linear SPM kernel on histograms and even better than the nonlinear SPM kernels.