Guard Alarm Device Based on Intelligent Object Recognition System
Download as PDF
DOI: 10.23977/amce.2019.015
Author(s)
Hang Zhou, Wei Dong, and Yinghui Jiang
Corresponding Author
Hang Zhou
ABSTRACT
At present, for major takeaways and pets entering the campus, the practice of major universities is to arrange doormen to block them at the school gate, or to send permits to the takeaways to enter the campus. To reduce the workload of doormen and improve their work efficiency, this paper developed a system based on artificial intelligence that can recognize objects, which can be combined with the program installed in the camera system at the entrance of the campus. When a deliveryman enters the campus, a prompt will sound in the doorman's room. This paper is based on the image processing technology to do the following content, first, use OpenCV haar cascade way to identify the takeaway workers into the campus, and issued an alert. Secondly, when the haar cascade mode of opencv was not accurate enough, we used Yolo to label the pictures and train our own customized "takeaway worker" training set and identify the target. The results show that: this paper has achieved a specific feature of the target, we can achieve accurate identification and take the corresponding action, reduce the workload of the guard.
KEYWORDS
Yolo, image processing, campus, takeaway, the guard