An Infrared Pedestrians Image Segmentation Algorithm
Qin Dai · 2013
In order to address the problem that the pedestrian segmentation in infrared image is easy to be interfered by the human pose and noise, this paper presents a pedestrian segmentation algorithm in infrared images employing super pixel and conditional random filed. Owing to accelerate the computation, the algorithm employs the simple linear iterative clustering algorithm to divide the image into some super pixels firstly. So as to represent the posterior of a distribution of the image, CRF model is applied to depict the configuration of super pixels in image. Finally we select the label in all of the possible states as the new label which has the minimum energy value in CRF model.