Gesture Recognition Applications Developed in Embedded Systems
| Autor: | Naixue Xiong, Shi-Jinn Horng, Xuan-Zi Hu, Wen-Lian Cao |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
business.industry
Computer science 02 engineering and technology Tracking (particle physics) Histogram of oriented gradients Ratio method Gesture recognition Embedded system Histogram 0202 electrical engineering electronic engineering information engineering Projection method 020201 artificial intelligence & image processing business Temporal difference learning Gesture |
| Zdroj: | PAAP |
| DOI: | 10.1109/paap.2018.00035 |
| Popis: | In this paper, the gesture recognition technique is developing in ARM embedded system. It is convenient, time-saving and low-cost effectiveness to use gestures to control machines. General gesture recognition technique is divided into three major parts: hand detection, hand tracking and gesture recognition. First the similar ratio method and Harris detection method are proposed for the hand detection. Then, the hybrid of Histogram of Oriented Gradients (HOG) and Mean-shift approach is used to hand tracking. Finally, the temporal difference method and scanning projection method do the gesture recognition. According to the results of the experiment, the proposed method has 89% correction rate for gesture recognition. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|