A robust method for counting people in complex indoor spaces

Abstract

People counting systems have wide potential applications including video surveillance and public resources management. This paper describes a robust method for counting people in complex indoor spaces. The method has counted the number of people using a single camera in the indoor spaces through four modules: image pre-processing module, morphology processing module, image marking module and people counting module, in order to master the information of the indoor for increasing efficiency and utilization of building facilities. Through experiment comparing, image pre-processing module chooses image graying, background subtraction based on threshold, median filtering and threshold segmentation to eliminate background interference. The morphology processing module uses the improved erosion operation and the improved dilation operation to extract target feature. Then the following image marking module uses connected component detection algorithm, setting the object feature and shape judgment condition and marking object region. Finally, people counting module is used to count the number of people. Experimental results show that the robust method can perform accurate people counting. The method is easily realized and suitable for the complex small-scale indoor spaces.