Face Recognition Attendance System and Emotion Recognition using Kinect Motion Capture Data of Human Gaits

The face is a key part of the human body that uniquely identifies a person. Using facial characteristics as biometrics, a facial recognition system can be implemented. The most challenging task in any organization is attendance marking. In the traditional attendance system, students are called by teachers and their presence or absence is marked accordingly. However, these traditional techniques are time-consuming and tedious. In this research paper, an open CV based face recognition approach was proposed. The training database is created by training the system with the faces of the authorized person. The cropped images are then saved as a database with the appropriate labels. Automatic emotion recognition has great value in many applications, but more portable, nonintrusive and low-cost technologies need to be developed to fully realize the application value of recognition. Human gaits could reflect the emotional state of the walker and could be a source of information for emotion recognition. This work proposed a new method of emotional state recognition through human walking using Microsoft Kinect, a low-cost, portable camera-based sensor. The walking of 59 participants in neutral condition, induced anger and induced happiness were recorded by two Kinect cameras, and the original data were processed by joint selection, coordinate system transformation, Gaussian sliding window filtering, differential operation and data segmentation. Features of gait patterns were extracted from the 3-dimensional coordinates of the 14 joints of the main body using Fourier transform and principal component analysis (PCA). NaiveBayes, RandomForests, LibSVM, and Sequential Minimal Optimization (SMO) classifiers were trained and evaluated, and the recognition accuracy of anger and happiness from the neutral state reached 80.5% and 75.4%, respectively. Although the results of distinguishing between anger and happiness states were not ideal in the current study, it demonstrated the feasibility of automatic recognition of emotional states from walking, while the features meet the application requirements.IJSRED - International Journal of Scientific Research and Engineering Developmen