Foot Recognition using LBP-kNN for Knee Rehabilitation

Autor: Jermphiphut Jaruenpunyasak, Pakpoom Hoyingcharoen, Mongkol Saejia, Rakkrit Duangsoithong
Rok vydání: 2018
Předmět:
Zdroj: 2018 15th International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON).
DOI: 10.1109/ecticon.2018.8619876
Popis: The foot recognition can be used in many applications such as gait pattern and knee monitoring exercises; nonetheless, the environment condition and a variety of foot are the main problems of the recognition. In general, the foot recognition system is designed in a closed room or a static background to recognize with bare foot in the same view but it is inconvenience for knee monitoring exercises for a patient. This paper presents a method for the foot recognition using LBP features with kNN classifier. The results are evaluated by accuracy and complexity. These results are also compared with HOG algorithm and SVM classifier. According to the results,the LBP-KNN techniques provides higher accuracy but higher complexity than other methods to recognize the foot from human image.
Databáze: OpenAIRE