Fig. 2From: A simplified nonlinear regression method for human height estimation in video surveillanceAn example of the proposed camera calibration system. a A sample video was taken, and some head and foot points were manually marked. b The camera used for collecting the dataset. c A scatter plot of the y-coordinate of observed and estimated head points with respect to the observed foot points. The initial parameters f=720, θ=−30, and c=300 are approximated by a visual estimation of the installed camera in (b), and optimal parameters are found as f=547.7, θ=−38.6, and c=270.2 based on the nonlinear regression methodBack to article page