Human body abnormal behavior detection method

Abstract

The invention relates to a human body abnormal behavior detection method. The method includes the steps of collecting video sequences of four behaviors including walking, running, punching and fallingof a person under a single fixed background, extracting a moving human body target through a Kalman filter algorithm, extracting an image of the moving human body target, storing the detected image of the moving human body target, constructing a training data set, extracting the Hu invariant moment, image entropy and aspect ratio feature of the image to establish a Bayesian classifier, and thus realizing classified recognition of four abnormal behaviors of the human body. The recognition rate and the real-time performance are great. The Bayesian classifier and a convolutional neural network method are combined to determine the final abnormal behavior class, so the recognition accuracy can be increased and misjudgment and missed detection can be reduced. The method can achieve detection and classified recognition of four abnormal behaviors including walking, running, punching and falling of a person under a fixed background. Not limited to the four behaviors, the method also can be used for judging other abnormal behaviors such as crowd gathering, fighting and the like.

Description

technical field [0001] The invention relates to a human body recognition technology, in particular to a human body abnormal behavior detection method based on a Bayesian classifier and a convolutional neural network. Background technique [0002] Abnormal behavior analysis technology has a wide range of applications in public security, smart home and other fields. Monitor the fall and coma of the elderly living alone, and monitor abnormal situations such as falling to the ground, fighting, abnormal gathering of crowds, and riots in crowded places. [0003] The current security method is to install cameras in public areas, but there are two main problems in the traditional monitoring mode: first, a large number of monitors are required to view the video around the clock, which can easily cause sensory fatigue, resulting in missed and wrong detections; second, Multiple cameras shot 24 hours a day without interruption, generating a large amount of data, which made it difficult...

AI Technical Summary

Problems solved by technology

[0003] The current security method is to install cameras in public areas, but there are two main problems in the traditional monitoring mode: first, a large number of monitors are required to view the video around the clock, which can easily cause sensory fatigue, resulting in missed and wrong detections; second, 24-hour uninterrupted shooting by multiple cameras generates a large amount of data, which makes it difficult to find later

[0004] In real scenes, there are often some uncontrollable factors, such as lighting changes, shadows, occlusions, viewing angle conversions, etc., which make the existing abnormal behavior detection methods very limited in practical applications. Therefore, it is necessary to improve the robustness of abnormal behavior analysis. Sexuality is an urgent issue that needs to be addressed

Images