Design and Implementation of a Machine Vision System for Standing Human Detection in Front of Tractor

These years, the use of tractors and self-propelled agricultural machines has been caused to increase the human death numbers due to accidents with them. Hence, having an alarm system on agricultural vehicles for humans and obstacles detection can be beneficial to reduce the accidents. This study was aimed to develop a machine vision system for human detection to use in agricultural tractors. A camera was installed in front of the tractor, and an algorithm was then designed in MATLAB for image processing. After trimming the input image, the human was identified using watershed and gradient magnitude techniques, and also subtracting human from the background. According to the results of the algorithm evaluation in various image acquisitions, the proposed algorithm had high accuracy at distances of 4 to 20 meters compared with two common human detection methods (Histogram of oriented gradient and Cascade). Average accuracy of the proposed algorithm in these distances was within 85 - 94 %. The results of variance analysis for accuracy of human detection showed that the effect of identification method and distance on the accuracy of human detection were significant as well as the interaction effects between different parameters were significant at the 1% level.