HSPOG: An optimized target recognition method based on histogram of spatial pyramid oriented gradients

The Histograms of Oriented Gradients (HOG) can produce good results in an image target recognition mission, but it requires the same size images for classification inputs. In response to this shortcoming, paper performs spatial pyramid segmentation on any size, gets pixel each block dynamically, and further calculates normalizes gradient oriented feature region layer. new is called Histogram Spatial Pyramid (HSPOG). This approach obtain stable vectors increase detection rate process significantly. Finally, article verifies algorithm using VOC2012 data compares effect HOG.