SVM-based Human Cell Detection Technique using Histograms of Oriented Gradients

An automatic human cell detection system is proposed in this article. Histograms of Oriented Gradients are used for cell feature extraction. A robust search procedure, using variable sized sliding windows, is performed for cell localization in grayscale images. The proposed sliding-window based search algorithm is used in combination with a non-linear SVM-based feature vector classification technique. Some successful experiments are described in this paper. Key-Words: object-class detection, cell detection, HOG features, window-based search algorithm, SVM classifier