KR20160072678A

KR20160072678A - Real time monitoring system for prevention of crime and violence

Info

Publication number: KR20160072678A
Application number: KR1020140180643A
Authority: KR
Inventors: 김홍성
Original assignee: 김홍성
Priority date: 2014-12-15
Filing date: 2014-12-15
Publication date: 2016-06-23

Abstract

In order to cope with anticipated violence or crime that may occur in a blind spot (closed space, unoccupied space, toilet, vacant lot, playground), CCTV can not be installed in terms of privacy. It is the reality that we are neglecting the crime.
The present invention relates to a system relating to violence and crime occurrence, discrimination, and processing having features that can be obtained while protecting the privacy of the on-site information that is generated or expected to occur in a blind spot.
According to the present invention, the motion information is obtained by extracting contour lines on the scene so as to discriminate violent and criminal situations while processing voice information and who is unknown in violence and crime scene, Extracts contour lines related to violence and criminal activity stored in the video and similar action video value and voice information discrimination reference value based on the information obtained in real time is compared with the emergency situation is determined, Extracts the contents stored in the voice tone from the storage device to induce the violation and the criminal act stop by warning broadcasting, stores the contents of the emergency situation (violence and crime-related outline-extracted moving picture and voice information) in the storage device, The administrator and the competent department are responsible for determining the emergency situation. Cognitive mobilization and violence that characterized in that to cope with emergencies and crimes in the development field, the present invention relates to the determination, the processing related systems.

Description

REAL TIME MONITORING SYSTEM FOR PREVENTION OF CRIME AND VIOLENCE

The present invention relates to a system for detecting and responding to violence and crime by using an infrared sensor, and more particularly, to a system for extracting motion-related outlines determined to be violent and criminal situations, It is determined that the information obtained on the basis of the extracted image of the outline and the urgent situation voice as the criterion of the stored violence situation is violence and crime situation, warning broadcast on the violence and crime scene, and the above contents are stored in the storage device And to a system capable of preventing and responding to violence and crime that is transmitted to an administrator or a department in charge so that a manager or a person in charge of the department can be dispatched to the field to release the emergency.

Techniques have been developed in which a camera is installed in a place that is open to the public and a manager (person) judges an emergency situation in real time and dispatches it to the site or stores it in a storage device.

However, there are few ways to cope with emergency situations, because emergency cameras are not installed due to privacy regulations in places where violence and crime are mainly occurred and places where people are uncommon and emergency bells are installed or left unattended.

In the meantime, it is impossible to install a device capable of obtaining information on violence and crime scene such as a camera in a blind spot (a closed space, a rare space, a toilet, an empty space, a playground, etc.) In order to solve the problem that the criminal act is neglected, it is necessary to obtain information on the situation of violence and criminal activity while protecting the privacy, so as to be able to prevent and cope with violence.

The present invention relates to a method and apparatus for extracting motion-related outline video so as to prevent discrimination of information on violence and crime scene while protecting privacy, acquiring voice information such as moaning, harshness, and rescue voice generated in the field, Extraction of contours related to violence and criminal activity It compares with the patterns of video, voice, and words to automatically identify the emergency situation. Then, it alerts to the scene and induces the violence to stop. At the same time, the manager or the person in charge of the department should check the information on the site and send them to the scene to end violence and crime.

According to the present invention, violence and crime scene information, which are mainly generated in blind areas, can not be obtained from the viewpoint of protecting the privacy, and it is possible to confirm movement related motion in the field without knowing who is on the scene in order to solve real problems in which violence and crime are neglected And automatically obtains voice information to inform emergency situation, etc., and terminates the situation through violence and crime prevention broadcasting and on-site dispatch.

FIG. 1 is a flow chart relating to violence and crime occurrence, discrimination, and coping for implementing the present invention.
Fig. 2 is an example of the image after extracting the contour of the scene figure and the image after extracting the contour line.

In order to obtain information on the situation of violence and crime scene occurring in public places, it is necessary to obtain the video of the scene in real time, or to listen to urgent voices (language, Moans, etc.) are analyzed and judged manually by humans.

In the present invention, attention is paid to the fact that violence and crime such as blindness are mainly generated in a place where privacy protection is required. In order to protect privacy and obtain information of violence and crime scene, an infrared sensor is used to detect violence and crime The image of the incoming light to the image sensor is transformed into a digital signal and the light coming into the image sensor is converted into a digital signal And extract the outline using the converted digital signal and output it to the screen.

In order to obtain motion information (first step 100 in Fig. 1), either far-infrared or near infrared rays can be applied. When using the infrared method, the distance of the light can be a problem, so you can add the function of illuminating the infrared light.

In the contour extraction (second step 101 in FIG. 1), a motion compensation method may be adopted to obtain an accurate moving image.

Of course, the contour (FIG. 2, 201) makes it possible to distinguish between violence and criminal activity while preventing the person from knowing for privacy.

Also, an amplification circuit may be added to the primary amplification or the secondary amplification for doubling the weak output signal of the infrared sensor.

The outline video of the violence and the criminal activity thus acquired is obtained by repeatedly learning the video of the outline about the violence and the criminal activity and then compared with the discrimination reference value stored in the database (the third step 102 in FIG. 1) .

Of course, the outline video of the violence and criminal activity stored in the database is prepared to discriminate the similar behavior pattern. In other words, in order to minimize the error of the criterion value, the video storing contour extraction video related to specific violence and criminal activity and the contour extraction video related to violent and crime similar behavior pattern are stored in the storage device.

On the other hand, voice information is obtained to recognize violence and crime scene emergency. (The first step 100 in FIG. 1) of the voice information (the second step 101 in FIG. 1) is obtained by receiving the voice information (high voice, moans, rescue request voice, etc.) do. Of course, an amplification circuit can be added to improve the sensitivity.

The voice information thus obtained is compared with a value for judging that the violence stored in the database (the third step (102) of FIG. 1) and the emergency situation of the crime scene are secured after a lot of trial and error.

As described above, although the infrared sensor is used in the present method, it is also known that the laser method can be applied to obtain violence and crime scene information. The infrared sensor application method adopted in the present invention has the advantage that it can be applied to a dark place with little light.

If the outline extraction video and audio information obtained as described above are compared with the database (the third step 102 in FIG. 1) and the emergency situation (the fourth step 104 in FIG. 1) is not determined, And returns to the acquisition (step 1 (100) in FIG. 1).

If it is determined to be in an emergency situation, violence and crime prevention can be

Warning broadcast (the fifth step 104 in Fig. 1) is exited. The content of the warning broadcast is determined by the department of the government or the public and the tone of the voice for broadcasting and broadcasting is stored in advance in the storage device and the address of the storage device storing contents suited to the situation of violence and crime scene is designated, To be broadcast on-site. Speakers can be installed one or more in one place.

The outline extraction video and audio information determined to be in an emergency state is recognized by a manager who manages violence and crime scene or a preset number preset in a jurisdiction department (police box, police station, guard company, etc.) (step 6, step 105 in FIG. 1) And the above contents are automatically stored in real time in the storage device of the manager and the competent department (FIG. 1, step (106)).

The administrator and the competent department can confirm whether the violent crime scene and the crime scene have been transmitted through the unique number or the location information (step 8 of FIG. 1) If it is determined that the situation is not an emergency, the file is deleted (step 10 of FIG. 1), and the process returns to the first step 100 of FIG. 1).

On the other hand, if it is determined that the situation is an emergency, the user is dispatched to the scene of violence and crime for the release of the emergency (Step 11 of FIG. 1 (110)). Violence and crime When a manager arrives at a crime scene and a person in charge of a department in charge of the crime scene,

100 ~ motion information or voice information obtaining step
101 ~ Motion contour extraction Video and audio information analysis phase
102 ~ Comparison with database (outline video and audio) values Step
103 ~ Emergency situation determination step
104 ~ Warning broadcasting phase
105 ~ Extracting outline and transmitting voice information to administrator and competent department
106 ~ Real-time storage step to administrator and jurisdictional department storage
107 ~ Steps to check the on-site address
108 ~ Emergency situation determination in managerial and competent department
109 ~ File deletion step
110 ~ Field trip stage
200 ~ Example video before contour extraction
201 ~ Example of video after extracting contour

Claims

(Daytime, nighttime) on-site information while protecting privacy to prevent or respond to violence or crime that may occur in, or occur in, blind spots (closed spaces, unoccupied spaces, toilets, The present invention relates to a method and apparatus for detecting violence and crime, which is characterized in that an emergency scene (violence and crime) can be released by securing an outline extraction video using an infrared sensor, obtaining voice information to discriminate an emergency situation, Occurrence awareness, discrimination and processing related system

Claim 1 of the present invention provides a method for obtaining information on violence and crime scene in real time by using a microphone and an infrared sensor to secure voice information (rescue request, moaning, A system that is capable of

The information obtained as described in claim 2 is compared with a value of an emergency situation judgment (violence and crime behavior judgment reference value and a pattern related to a reference value such as a structure request, a moaning sound, &Lt; RTI ID = 0.0 >

The system according to claim 3, further comprising a function of selecting a content and a voice tone suitable for the scene of the contents stored in the storage device based on the determination of the emergency situation,

(Outline video and audio information) to the responsible department such as the manager and the department in charge (police box, police station, guard company, etc.) on the basis of the emergency situation determined as described in claim 3 and automatically stores the emergency situation in the storage device &Lt; RTI ID = 0.0 >

Characterized in that it has a function of continuously repeating claims 2 to 5 in real time