Human Detection in Images Based on Histogram of Oriented Gradients with evaluating the DENON criteria
عنوان مقاله: Human Detection in Images Based on Histogram of Oriented Gradients with evaluating the DENON criteria
شناسه ملی مقاله: ITCT13_016
منتشر شده در سیزدهمین کنفرانس بین المللی فناوری اطلاعات،کامپیوتر و مخابرات در سال 1400
شناسه ملی مقاله: ITCT13_016
منتشر شده در سیزدهمین کنفرانس بین المللی فناوری اطلاعات،کامپیوتر و مخابرات در سال 1400
مشخصات نویسندگان مقاله:
Ayeh Ansari - Faculty of Biomedical Engineering, Khorasan Institute of Higher Education, Mashhad, Iran
Mohammad Hasan Olyaei Torqabeh - Faculty of Electrical Engineering, Sadjad University of Technology, Mashhad, Iran
خلاصه مقاله:
Ayeh Ansari - Faculty of Biomedical Engineering, Khorasan Institute of Higher Education, Mashhad, Iran
Mohammad Hasan Olyaei Torqabeh - Faculty of Electrical Engineering, Sadjad University of Technology, Mashhad, Iran
This paper discusses the Histogram of Oriented Gradients (HOG) algorithm first. Then, we implemented this algorithm using the python language and the OpenCV library. The results show that this algorithm has sufficient ability to human detection in images. For a closer look, according to the DENON criteria, we implemented this algorithm on ۱۰۰۰ images of the MPII database and we calculated the accuracy of this algorithm. The results show that the HOG algorithm correctly detects ۸۰۰ images from a total of ۱۰۰۰ images, and obtains an accuracy of ۸۰%.
کلمات کلیدی: Histogram of Oriented Gradients, HOG, DENON, python, OpenCV
صفحه اختصاصی مقاله و دریافت فایل کامل: https://civilica.com/doc/1326416/