| Title |
Intelligent Digital Signage System Implementation based on Emotion Recognition Algorithm |
| Authors |
최세종(Se-Jong Choi) ; 조연희(YeonHee Jo) ; 손일수(Illsoo Sohn) |
| DOI |
https://doi.org/10.5573/ieie.2019.56.3.63 |
| Keywords |
digital signage ; intelligent advertisement system ; emotion recognition ; graph theory |
| Abstract |
Recently, a variety of intelligent digital signage is emerging due to the fusion of digital signage and ICT. Intelligent digital signage enhances the effectiveness of advertisements and meets the needs of both advertisers and consumers. There is also a new revenue generating model based on this. In this paper, we design and implement an intelligent digital signage system that maximizes the advertising effect. The proposed intelligent digital signage system collects the time-based emotion information of the consumers who watch the advertisement. It maximizes the advertisement effect by scheduling the advertisements with high emotion-advertisement correlation factors. The proposed system consists of five functional blocks: Video data collection, Face detection, Emotion recognition, Advertisement scheduling and Advertisement display. Various artificial intelligence algorithms based on OpenCV are used. In particular, in the advertisement scheduling stage, we formulate a matching problem to maximize an advertisement utility based on a graph theory and develop a novel matching algorithm to find the optimal solution. The proposed intelligent digital signage system is implemented as a general purpose computer platform equipped with a camera and a display hardware, and can be effectively used in commercial advertisement systems. |