Humankind shape object detection using deformable parts model with HOG features

It is one of the important tasks in machine vision field to detect specific object using single image. A machine learning approach using latent support vector machine(LSVM) classifier was presented for detecting humankind shape object. The histogram of oriented gradients(HOG) features were extracted and the corresponding deformable parts were formed to describe the appearance of the object. The problem that the objects appearance deformed when it was in motion, was solved. 200 images captured in typical public areas were randomly selected and used to perform this method. 1 100 humankind shape objects from them were tested, and the correct recognition rate of 78.3% was achieved. The experimental results show that this approach is able to detect humankind shapes and mark them out which demonstrate its feasibility and stability;however,misdetection can happen when the humankind shape object is partly covered.