Device for automatically recognizing human body malignant violent terrorism actions and emitting violence preventing substances

Abstract

The invention relates to a device for automatically recognizing human body malignant violent terrorism actions and launching violence preventing substances. The device comprises an intelligent recognizing terminal and electric executing mechanism for automatically launching stun grenades. The intelligent recognizing terminal comprises an infrared camera which measures space field depth data on the basis of the structure light measuring technology, an X86 industrial personal computer and a violence preventing control device. The infrared camera and the violence preventing control device are connected to the X86 industrial personal computer. A violent action comparison library and a violent instrument comparison library are further disposed in the X86 industrial personal computer. The intelligent recognizing terminal can recognize high-danger actions of knife slashing and armed violent attack in real time and uses an electric control percussion type WJ9201 type 38mm violence preventing launcher which is developed by the Zhejiang Xinhua machine factory to automatically launch the 38mm stun grenades so as to stop continuous deterioration of the violent terrorism behaviors. The device can effectively automatically recognize the high-danger actions and launch the violence preventing substances according to needs, and the continuous deterioration of violent terrorism events can be stopped timely.

Description

technical field

[0001] The invention relates to a device for automatically identifying vicious acts of human violence and launching explosion-proof substances, belonging to the technical field of 3D visual perception and recognition of human limb movements. Background technique

[0002] Due to technical limitations, all current security systems are based on 2D vision plus strong algorithms, but the world we see is three-dimensional. Based on 2D vision plus strong algorithms, the image analysis error is large and the accuracy is poor. , resulting in violent actions on the human body and dangerous control knives in the monitoring scene, the problem can only be found by the remote manual identification of the monitoring screen by the security personnel. Even if the security personnel can find the problem through the monitoring screen, they cannot go to the scene to stop and deal with it in time High-risk incidents on the scene. Especially in some special occasions, the monitor...

Images