Gesture recognition has become increasingly popular, in response to the growing demand for intelligent and personalized human-computer interaction (HCI) and human-to-human interaction. However, gesture recognition raises a high requirement on the background color of the gesture image, and faces difficulty in extracting multiple gesture features. To solve these problems, this paper presents a novel approach for gesture feature extraction and recognition based on image processing. Firstly, the workflow of the proposed gesture recognition method was given, and a series of preprocessing was performed on the original gesture image, prior to formal extraction and recognition. Next, the authors detailed the extraction of features from gesture boundaries and fingertips. Finally, a convolutional neural network (CNN) was constructed for gesture recognition, and a gesture recognition model was developed based on residual network. The proposed approach was proved to be valid through experiments. The research results provide a reference for the application of CNN in the recognition of various postures or shapes.

No abstract

The traditional solving algorithms for human attitude face problems like poor stability and low accuracy. To overcome these problems, this paper puts forward a novel human attitude solving algorithm based on fuzzy proportional-integral-derivative (PID) controller and complementary filter, which integrates the data collected by accelerometer, magnetometer and gyro. Through complementary filtering, the error of the gyro was corrected with the aid of accelerometer and magnetometer. Based on the traditional proportional-integral (PI) controller, the fuzzy control was introduced to adjust the parameters in real time, and the differential control (D) was added to improve the dynamic performance of the system, creating a fuzzy PID controller. Then, the fuzzy PID controller was adopted to control the complementary filtering, and the quaternion updating method was employed to compute the human attitude. To verify its effectiveness, our algorithm was compared with the traditional PI filtering algorithm through static and dynamic experiments, using an MPU9150 nine-axis motion tracking device and the MATLAB on the upper computer. The experimental results show that our algorithm achieved stable and accurate output of attitude angles. The research findings are of great application potential in rehabilitation therapy, virtual reality (VR) and human-computer interaction (HCI).