Depth image tumbling recognition algorithm based on wavelet moment

Abstract

The invention provides a depth image tumbling recognition algorithm based on a wavelet moment. The algorithm comprises the steps of firstly normalizing the distances between all pixel points in an image and the centroid, then conducting coordinate and FFT transformation on the normalized image, finally conducting wavelet transformation on the image to obtain a characteristic vector and recognizinghuman body behaviors in combination with a minimum distance classification method. By testing a large quantity of human body behavior samples, the recognition rate of the algorithm reaches 90%. Through a depth image and translation, scaling and rotation invariance of the wavelet moment characteristics, the privacy problem which exists in traditional video image processing is solved, and the recognition performance is improved. The recognition capability of the human body tumbling behaviors is improved, and the algorithm has better robustness and a very good application prospect.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and in particular relates to a wavelet moment-based depth image fall recognition algorithm. Background technique [0002] Fall detection is a hot issue in human abnormal behavior detection, and many researchers at home and abroad have carried out research on it. At present, there are mainly the following methods: (1) Based on wearable devices [Document 1] (P.Pierleoni, A.Belli, L.Palma, M.Pellegrini, L.Pernini, S.Valenti.A high reliability wearable device forelderly fall detection[J].IEEE Sensor.2015,15(5):4544-4553.): Using sensors such as accelerometers to detect the motion parameters of objects, the comfort and scalability are poor. (2) Based on audio frequency information [Document 2] (Litvak D, Zigel Y, Gannot I. Fall Detection of Elderly through Floor Vibrations and Sound [C]. Engineering in Medicine and Biology Society, International Conference of the IEEE.2008:4632-4635 .): This...

AI Technical Summary

Problems solved by technology

At present, there are mainly the following methods: (1) Based on wearable devices [Document 1] (P.Pierleoni, A.Belli, L.Palma, M.Pellegrini, L.Pernini, S.Valenti.A high reliability wearable device forelderly fall detection[J].IEEE Sensor.2015,15(5):4544-4553.): Using sensors such as accelerometers to detect motion parameters of objects, poor comfort and scalability

(2) Based on audio frequency information [Document 2] (Litvak D, Zigel Y, Gannot I. Fall Detection of Elderly through Floor Vibrations and Sound [C]. Engineering in Medicine and Biology Society, International Conference of the IEEE.2008:4632-4635 .): This method uses the vibration frequency signal generated by the collision between the body and the ground when falling to judge the fall of the human body, but it is susceptible to noise interference

(3) Based on 2D video image capture [Document 3] (Maldonado C, Ríos H, Mezura-Montes E, et al. Feature selection to detect fallen pose using depth images [C]. International Conference on Electronics, Communications and Computers. IEEE, 2016: 94-100.): Image processing technology is used to detect the object information obtained by the video surveillance system. However, the delay of image transmission leads to limited real-time performance of behavior detection, and the detection effect drops sharply under insufficient light and shadows.

At present, relevant research is in its infancy, and there is a lot of room for research

Images