Detection method for body conflict event

Abstract

The invention relates to dynamic image processing, in particular to a detection method for a body conflict event. The detection method comprises the following steps of: filtering an image in a relative static state by frame difference information of an image sequence to obtain profiles of motion objects; determining the quantity and the independence of the motion objects by carrying out filteringand noise reducing and then using an image projecting method; if the quantity of the motion objects is smaller than 2 or the motion objects are not crossed, showing that the body conflict event does not exist and otherwise carrying out accurate detection of the object bodies on the image, i.e. carrying out bidirectional motion estimation on the motion objects by using a block matching method; extracting a characteristic quantity of each motion object and constructing a behavior judgment model; and when the bidirectional motion characteristic quantity of at least one motion object respectivelymeets the judgment rule, showing the existence of the body conflict event and otherwise showing the inexistence of the body conflict event. According to the method disclosed by the invention, unattended condition and conflict event alarm in a monitoring area can be realized.

Description

technical field [0001] The invention relates to image processing, in particular to the field of dynamic image analysis and processing, and in particular to a detection method for body conflict events. Background technique [0002] With the development of computer technology, communication technology and image processing technology, video surveillance has been widely used in various applications. However, in most cases at present, the video surveillance terminal still needs to be on duty manually, and the relevant information of the monitoring site can be learned by switching the video surveillance screen regularly or manually according to the needs, so as to judge whether a conflict has occurred. Obviously, the manual on-duty method has shortcomings such as huge workload, slow response to conflict events, and missed detection and reporting of conflict events. [0003] The introduction of video image processing technology is one of the effective ways to solve the above probl...

AI Technical Summary

Problems solved by technology

On the one hand, its human foreground object extraction module uses the Codebook algorithm to extract foreground human objects. However, the target objects extracted by this algorithm include static and moving human objects, but it does not specifically target violent behaviors that cause human body (especially limb) movements. Due to the feature of large amplitude, the corresponding image filtering method is used to filter out the image of the human body in a relatively static state. Under normal circumstances, the human body in the elevator is generally in a relatively static state or occasionally has a short-term movement. The object extraction module needs to process a large number of image sequences, so its processing speed is not high; on the other hand, it needs to go through two stages of training and detection to realize the violent behavior detection in the elevator by using the hidden Markov model (HMM)

In the training phase, it is necessary to obtain a large amount of eigenvector data of normal behavior in the elevator. The time required for training is bound to be affected by the size of the training sample. The larger the sample, the longer the training time, so the detection speed is restricted by the training time.

In addition, this patent only analyzes the existing video images based on the relevant features of the human body object in the video image as the foreground connected area in the violent behavior, and does not make further predictions on the current behavior, so as to more accurately determine whether it is violence Things, like this, have the potential to misjudge well-meaning, relatively quick physical contact between characters as violence

Images