KR101535890B1 - Image monitoring method using text recognition and apparatus thereof - Google Patents

Abstract

The present invention relates to a method and a device for monitoring an image by using text recognition. According to the present invention, a method for monitoring an image of a device for monitoring an image by using text recognition comprises the steps of: detecting at least one trade name region from a photographed image photographed through a camera; recognizing text information existing in the trade name region; storing at least one among camera control information set by matching with the text information, the text information, and position information of the camera; and controlling the camera based on the camera control information to photograph a standard region of a screen corresponding to the trade name region when a request for an image including the text information is received. According to the method and device for monitoring an image by using text recognition, text information is recognized from a trade name around the device for monitoring an image. When a request for an image for a particular trade name or a phone number is received in the future based on a camera setting value set by matching with the recognized text information, a photographing image for a point corresponding thereto is expected to be rapidly and actively provided.

Description

TECHNICAL FIELD [0001] The present invention relates to an image monitoring method using text recognition,

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video surveillance method and apparatus using text recognition, and more particularly, to a video surveillance method and apparatus using text recognition that can increase the efficiency of video surveillance.

Video surveillance systems using CCTV cameras are installed in places such as buildings, roads, alleys, parks, and indoor spaces, and are used for the purpose of methods, field surveillance, various crime solutions, and crime prevention.

The images captured by the video surveillance device are transmitted to the operation server in real time and can be checked in real time on the operation server. Also, by controlling pan / tilt of the camera on the operation server, the environment of video surveillance can be controlled.

Conventionally, when an incident such as a specific accident or a crime occurs and urgent help is needed, the video surveillance apparatus searches for a video surveillance device and performs image verification on the address depending on the address information reported by the notifier. However, such a method has a problem in that it is difficult to promptly search an image of an incident site if the applicant does not know the specific address information.

Furthermore, when the address information is received from the complainant, the operator must manually adjust the angle of the video surveillance device at the address and manually search the incident site until the accident point or the claimant is exposed on the screen. Such a method has a problem that it is difficult to make a field check process considerably and it is difficult to quickly judge and respond to an event.

Also When a police officer is put into a crackdown on a specific site (ex, PC room, or entertainment agency), people may be aware of it and escape in advance. As a method for capturing these images, the operator of the server must manually set the angle of the camera and prepare the images of the camera installed in the place before the police input. However, this setting and setting process is cumbersome and inefficient.

The technology which is the background of the present invention is disclosed in Korean Patent Publication No. 2009-0096140 (published on September 10, 2009).

In accordance with the present invention, when a video request corresponding to a mutual point is received based on a previously stored camera setting value corresponding to a mutual region around the video surveillance apparatus, the captured image for the corresponding point is immediately and actively The present invention provides a video monitoring method using text recognition that can be provided.

The present invention relates to a video surveillance method for an image surveillance apparatus using text recognition, comprising the steps of: detecting at least one commodity area from an image captured through a camera; Storing at least one of camera control information set in matching with the text information, the text information, and the location information of the camera; And controlling the camera based on the camera control information so that a reference area of the screen corresponding to the mutual area is captured when the video request including the text information is received. to provide.

Here, the video surveillance method using text recognition may further include transmitting the stored information to an operation server, and the step of controlling the camera may include receiving a video request including the text information from the operation server , The image information captured by controlling the camera can be transmitted to the operation server.

In addition, upon receipt of the video request including the text information and the location information of the camera, the operation server may transmit the video request to the video surveillance apparatus including the camera of the location information among the plurality of video surveillance apparatuses .

The camera control information may include at least one of PAN, Tilt, and Zoom, and the location information of the camera may include Global Position System (GPS) information or address information. can do.

In addition, the video surveillance apparatus may separately store at least one of the camera control information, the text information, and the camera position information for the plurality of mutual areas to be detected and provide the operation server with the information.

Further, the mutual area includes a character area formed by a plate-shaped area or a character block, and the text information in the mutual area may include letters or numbers.

The step of recognizing the text information existing in the mutual area may include the step of recognizing the text information when at least one of a foreign language character, a special symbol and a numeral is included in the mutual area or if the text information is not recognized from the mutual area, Quot ;, or " modify "

According to another aspect of the present invention, there is provided a computer-readable storage medium storing a program for causing a computer to function as: a mutual area detection unit for detecting at least one commercial area from a captured image photographed through a camera; a recognition unit for recognizing text information existing in the mutual area; A storage unit configured to store at least one of camera control information, text information, and camera position information set by the camera control unit, and a display unit configured to receive a video request including the text information so that a reference area of a screen corresponding to the mutual area is captured And a control unit for controlling the camera based on the camera control information.

Here, the video monitoring apparatus using the text recognition may further include a communication unit for transmitting the stored information to the operation server and receiving a video request including the text information from the operation server, Upon receipt of the video request including the information, the controller can control the camera to transmit the captured image information to the operation server.

According to the present invention, when text information is recognized from a mutual vicinity of a video surveillance apparatus and a video request for a particular name or telephone number is received based on a camera setting value set to match the recognized text information, It is possible to expect an effect of quickly and actively providing a picked-up image of the picked-up image.

1 is a block diagram of an image monitoring apparatus using text recognition according to an embodiment of the present invention.
2 is a flowchart of a video surveillance method using FIG.
FIG. 3 is a conceptual diagram showing information storage according to recognition of mutual regions in the embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art can easily carry out the present invention.

The present invention relates to a video surveillance method and apparatus using text recognition, and more particularly, to a video surveillance method and apparatus using text recognition, When a video request corresponding to the mutual point is received, the user can promptly and actively provide the sensed image of the corresponding part without any additional operation by the user.

In the video surveillance method according to the embodiment of the present invention, when an incident such as a specific accident or a crime occurs, the name (eg, a business name or a telephone number written on the sign) , The camera of the video surveillance apparatus corresponding thereto is automatically driven in accordance with the camera control information (pan / tilt value of the camera) corresponding to each other, so that the video of the incident scene can be immediately confirmed in the optimal view .

1 is a block diagram of an image monitoring apparatus using text recognition according to an embodiment of the present invention. 1, an image monitoring apparatus 100 using text recognition according to an exemplary embodiment of the present invention includes a photographing unit 110, a cross area detecting unit 120, a recognizing unit 130, a storing unit 140, (150), and a control unit (160).

The photographing unit 110 photographs a surrounding image through a camera, and captures an image including a specific crossing area under the control of the control unit.

The inter-area detection unit 120 detects at least one common area from the image of the building obtained through the photographing unit 110. [ Such a mutual area may refer to a region where a signboard exists (hereinafter referred to as a signboard area) in an outer wall of a building or a surrounding area thereof.

The recognition unit 130 recognizes text information existing in the mutual area. Here, the text information includes a business name, a telephone number, and the like existing in the mutual area.

The storage unit 140 stores at least one of camera control information, text information, and camera position information matched with the text information. The camera control information may correspond to at least one of a PAN, a tilt, and a zoom of a camera for capturing a reference area of a screen corresponding to the mutual area. Of course, when the surrounding mutual area is photographed through the zoom, the zoom information can be stored together with the pan / tilt information.

The communication unit 150 transmits information stored in the storage unit 140 to the operation server 200. Here, the operation server 200 collects and collects the information received from the plurality of video surveillance apparatuses 100 that it manages.

The communication unit 150 may receive a video request including the mutual information from the operating server 200 or the operating terminal 300 (ex, smart phone, tablet, etc.).

Here, in the embodiment of the present invention, the operating terminal 300 accesses the operating server 200 to make a video request and receive a response thereto, but the operating terminal 300 directly connects to the video surveillance apparatus 100 And may receive the video request and receive a response thereto. In this case, it is preferable to include the function of the operation server 200 in the operating terminal 300.

Upon receiving the video request including the text information, the control unit 160 controls the camera based on the camera control information stored in the storage unit 140 so that the reference area of the screen corresponding to the mutual area is captured, And transmits one image information to the operation server 200 in real time.

The control unit 160 may receive the camera control (including Pan / Tilt / Zoom information) request from the operation server 200 or the operation terminal 300 to control the camera. If the PTZ value To store the mutual area.

According to the embodiment of the present invention, the video surveillance apparatus recognizes letters / numbers and the like related to neighboring locations and stores them together with the camera setting values (pan / tilt values) necessary for photographing the mutual points, When a video request is made at a specific point, the point can be photographed quickly and actively in an optimal field of view.

1 illustrates that the video surveillance apparatus 100 is provided outside the operation server 200. However, the video surveillance apparatus 100 according to the embodiment of the present invention is provided in the operation server 200 and operates It is possible.

Embodiments of the present invention will be described in more detail with reference to the above-described configuration. FIG. 2 is a flowchart of a video surveillance method using FIG. 1, and FIG. 3 is a conceptual diagram illustrating information storage according to recognition of mutual regions in an embodiment of the present invention.

As shown in FIG. 2, the inter-area detection unit 120 of the video surveillance apparatus of the present invention detects at least one common area from the captured image photographed through the photographing unit 110 (S210). Here, the mutual area may mean an area where a signboard exists among the outer wall of the building or the surrounding area as described above. The mutual area can be detected at least one per building depending on the number of the signboards mounted.

The mutual area includes an area of a plate-like signboard, which is generally made in a plate shape and mounted on an outer wall of a building, and an area in which a letter signboard that forms a single signboard is gathered one by one.

The letter type signboard may be in the form that an individual character block is mounted on the exterior wall of the building, or in a form attached separately to the glass surface. Here, in the case of a character type signboard other than a plate-like type, the inter-area detection unit 120 may determine object blocks adjacent to each other as a single signboard group and detect it as a single mutual area.

The types of signboards are variously disclosed, and a method of detecting a mutual existence area from a captured image of a building can be implemented by using various well-known object detection techniques and image processing techniques.

As described above, in the case of the plate-like signboard, the mutual area detecting unit 120 can detect the mutual area based on the rim of the signboard. Conventional edge detection techniques can be used in this process. In the case of the character type signboard, the cross area detecting unit 120 can detect, as a mutual area, an area in which any objects arranged with text attributes on the building are crowded.

Of course, the inter-area detection unit 120 may also be a potential inter-area where edges may be present through edge detection, such as an outer area (mainly a rectangular area) forming each layer of a building and a window area classified by the outer edge of the window It is also possible to detect it. As described above, there are many variations in the method of detecting the mutual region.

Next, the recognition unit 130 recognizes the text information existing in the mutual area (S220). Here, the text information may include letters or numbers. A business name (letter or number, or a combination thereof), telephone number (number), and the like may be described in the mutual area.

Since the mutual area is an area in which the text information (ex, business name, telephone number) corresponding to the mutual area exists, if the image processing or the like is performed from the detected mutual area, the corresponding text information can be recognized.

There are many techniques for detecting text or numbers from a captured image, so that recognizing or extracting text in an image can be selectively used among the known methods.

Next, the controller 160 stores at least one of the camera control information, the text information, and the camera position information matched with the text information in the storage unit 140 (S230).

For example, the control unit 160 may previously determine and store camera control information (camera setting value) for causing a reference area (e.g., a front gate of a building) of the screen corresponding to the detected inter-area (140).

The location information of the camera may include an address of a location where the actual camera is located, or the installation coordinates of the camera.

At this time, the address to be stored may correspond to a general address form such as the address of XXXXXXXXXXX of Seoul, the name of the building, and the like. In addition, the coordinate information (ex, actual GPS coordinates, block information of the corresponding zone) is additionally storable and may additionally be used for accurate location management of the camera.

Then, the communication unit 150 transmits the stored information to the operation server 200 (S240). The operation server 200 can receive the information from a plurality of video surveillance devices scattered in various areas and can integrally manage the received information. At this time, the operation server 200 may manage the information received from the plurality of video surveillance apparatuses 100 according to the region / sub-region / building attribute of each video surveillance apparatus.

Referring to FIG. 3, the inter-area detecting unit 120 detects a plurality of mutual areas on the building through the photographing unit 110. [ Then, the recognition unit 130 recognizes each of the mutual areas 'the 24 hour convenience store', 'the coffee', and 'the TOEIC institute' as text information. Of course, as shown in FIG. 3, the telephone number of the shop can be further recognized from the mutual area, but it will be omitted for convenience of explanation.

If the text information is not recognized from the mutual area, the recognizing unit 130 may recognize at least one of a foreign language character, a special symbol, and a number in the mutual area, Text may be entered or modified from the administrator.

That is, when the mutual region is composed of English, numbers, symbols, Chinese characters, etc., the recognition unit 130 receives input or correction of a national language corresponding to a business name commonly called a name (for example, "#" Quot;? &Quot;), text information recognized by the administrator is input from the manager in order to prepare for the case where characters in the mutual area are not recognized (for example, fonts and shapes of letters are not recognized uniquely) Or a function capable of being corrected.

3, the controller 160 acquires the pan / tilt values of the camera when the user is at the reference point of the screen corresponding to each mutual area, for each mutual area, and stores the pan / .

As described above, the video surveillance apparatus 100 may store the camera control information, the text information, and the camera position information corresponding to the detected plurality of mutual areas, and may provide the camera control information, the text information, and the camera position information to the operation server 200. Also, one video surveillance apparatus can individually store and transmit pan / tilt information for a plurality of signboards within the pan / tilt view of the user, and has the ability to cover various signboards.

In the case of FIG. 3, it is illustrated that one video surveillance device exists adjacent to a building for the sake of explanation, but there may exist a plurality of video surveillance devices sporadically. Accordingly, when multiple video surveillance apparatuses are used, the embodiments of the present invention can provide images at different angles with respect to one target, thereby enabling multi-angle image analysis when an event occurs.

If there are a plurality of video surveillance devices in the vicinity of the building as described above, the process shown in FIG. 3 may be performed for each video surveillance device. As an example, assume that there are two video surveillance devices for a building.

It is assumed that the first video surveillance device is capable of capturing all of the areas 1, 2, and 3 within the range of its own camera rotation range, and the second video surveillance device is capable of capturing only the areas 1 and 2. The first video surveillance apparatus recognizes the text information for the regions 1, 2, and 3 and individually acquires pan / tilt information corresponding to the regions 1, 2, and 3. In this way, the second video surveillance apparatus recognizes the text information for the areas 1 and 2, and separately acquires and stores the pan / tilting information corresponding to the areas 1 and 2.

When each of the video surveillance apparatuses is requested to shoot an image corresponding to the mutual image from the operation server 200, the respective video surveillance apparatuses can individually photograph the corresponding point and provide the shot to the operation server 200. [ Here, the region 1 or the region 2 is an area covered by both image sensing devices. In this case, the operation server 200 can receive two images corresponding to one mutual name, and it is possible to perform multi-view surveillance on one target point. Of course, the operation server 200 may selectively request some video surveillance devices among the plurality of video surveillance devices.

On the other hand, when receiving a video request including the mutual name from the operation server 200, the video surveillance apparatus is driven and controlled at a corresponding pan / tilt angle, To take a picture. Accordingly, a peripheral image including a signboard portion or a specific point (for example, a front door of a building, a window, etc.) is photographed.

More specifically, the communication unit 150 of the video surveillance apparatus 100 receives the video request including the ex (the coffee) from the operation server 200 or the operation terminal 300 (S250).

At this time, when receiving a video request, address information as well as mutual information can be received together. This is to prevent unnecessary operation of video surveillance systems at other points where the same mutual store exists in multiple locations or buildings.

As described above, in step S250, the operation server 200 can receive, from the user (manager), address information (ex, Seoul-dong) where the mutual corresponding text information (ex, . Then, the operation server 200 can transmit the video request to the video surveillance apparatus including the camera corresponding to the input position information of the plurality of video surveillance apparatuses 100.

At this time, the user may input information based on the information received from the field caller. For example, the operation server 200 confirms the mutual text information (ex, the name of the business name and telephone number of the sign) and the location information (simple address) visible in the field of the declarant at the place where the specific event occurs, And input it.

The embodiment of the present invention can use a simple address (eg, Seoul, Seoul) recognized by the case declarer or nearby facility information (ex, neighboring station, building name, landmark) . For example, if the operation server 200 receives the information of 'add-on station' and 'the coffee', the video surveillance apparatus corresponding to the coffee shop in the vicinity of the add-on station is searched and driven do.

In addition, when there are a plurality of corresponding video surveillance devices matching the text information and the positional information received from the user, the operation server 200 may display at least one (one or more) video surveillance devices among a plurality of video surveillance devices And transmits a video request including text information. At this time, the operation server 200 may automatically select an arbitrary one of the plurality of video surveillance apparatuses or may directly select the one video surveillance apparatus from the administrator.

Upon receiving the video request from the operation server 200, the control unit 160 controls the camera based on the camera control information so that the reference area of the screen corresponding to the mutual area is captured, and transmits the captured video information to the operation server 200 In real time (S260). Of course, the control unit 160 may control the camera so that the mutual region is picked up at a predetermined region of the screen (for example, the center of the screen or the upper end of the screen) to photograph the image and provide it to the operation server 200.

According to the embodiment of the present invention, when only the mutual information and the positional information are confirmed from the event notification, the image of the corresponding point can be immediately photographed and provided by using the preset camera pan / tilt angle corresponding thereto.

For example, when the complainant notifies the user of " Seoul Gasan-dong, the coffee ", the operation server 200 receives this information from the user, searches for a video surveillance device capable of photographing the area and the corresponding area, So that it is possible to capture images corresponding to each other. The video surveillance apparatus further captures a surrounding image of the coffee shop and provides it to the operation server 200.

The image sensing method of the video surveillance apparatus can automatically detect the accident point without having to manually adjust the pan / tilt angle by the operator, so that the on-site confirmation is easy and the case can be quickly judged and responded. Also, according to the embodiment of the present invention, even if the claimant does not know the detailed address of the scene of the incident, the incident site can be immediately searched and tracked based on the brief information of the area and the shop information.

According to the video surveillance method and apparatus using text recognition according to the present invention, text information is recognized in a mutual area around the video surveillance apparatus, and corresponding camera control information is stored in advance, When a video request for a telephone number is received, there is an advantage that a captured image for the corresponding point can be promptly and actively provided based on the camera control information corresponding thereto.

In addition, the present invention can be easily applied to the system without changing the structure of the existing method system, which greatly enhances the utilization of the crime prevention system in various parts of the country.

While the present invention has been described with reference to exemplary embodiments, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims. Accordingly, the true scope of the present invention should be determined by the technical idea of the appended claims.

100: video surveillance apparatus 110: cross-
120: recognition unit 130: storage unit
140: communication unit 150:
200: operating server 300: operating terminal

Claims (14)

A video surveillance method of a video surveillance apparatus using text recognition,
Detecting at least one commodity region from a sensed image photographed through a camera;
Recognizing text information existing in the mutual region;
Storing at least one of camera control information including at least one of pan, tilt, and zoom set in correspondence with the text information, the text information, and the position information of the camera; ; And
And controlling the camera so as to capture a reference region of a screen corresponding to the mutual region based on at least one of pan, tilt, and zoom when receiving a video request corresponding to the text information, The method comprising the steps of: The method according to claim 1,
And transmitting the stored information to an operation server,
Wherein the controlling the camera comprises:
And transmitting the captured image information to the operation server by controlling the camera upon receiving a video request corresponding to the text information from the operation server. The method of claim 2,
The operating server,
And receiving the video request corresponding to the text information and the camera position information, transmits the video request to the video surveillance apparatus including the camera of the position information among the plurality of video surveillance apparatuses. The method according to claim 2 or 3,
The location information of the camera may include:
GPS (Global Position System) information or address information. The method of claim 2,
The video surveillance apparatus includes:
And storing at least one of the camera control information, the text information, and the location information of the camera for each of the plurality of mutual areas to be detected, and providing the stored information to the operation server. The method according to claim 1,
Wherein the mutual area includes a character area formed by a plate-shaped area or a character block,
Wherein the text information in the mutual region includes text or numbers. The method according to claim 1,
Wherein the step of recognizing text information existing in the mutual area comprises:
Wherein the text information includes at least one of a foreign language character, a special symbol, and a number in the mutual area, or if the text information is not recognized from the mutual area, inputting or modifying the text information from the manager. A mutual area detection unit for detecting at least one commercial area from a captured image photographed through a camera;
A recognition unit for recognizing text information existing in the mutual area;
A storage unit for storing at least one of camera control information including at least one of PAN, tilt, and zoom set in correspondence with the text information, the text information, and the position information of the camera; part; And
And controlling the camera so as to capture a reference region of a screen corresponding to the mutual region based on at least one of pan, tilt, and zoom when receiving a video request corresponding to the text information, And a control unit for controlling the display unit to display the text. The method of claim 8,
Further comprising a communication unit for transmitting the stored information to the operation server and receiving a video request corresponding to the text information from the operation server,
Wherein,
And transmitting the captured image information to the operation server by controlling the camera upon receiving a video request corresponding to the text information. The method of claim 9,
The operating server,
And transmits the video request to the video surveillance apparatus including the camera of the position information among the plurality of video surveillance apparatuses when the video request corresponding to the text information and the camera position information is received. The method according to claim 9 or 10,
The location information of the camera may include:
A video surveillance apparatus using text recognition including GPS (Global Position System) information or address information. The method of claim 9,
The video surveillance apparatus includes:
And storing at least one of the camera control information, the text information, and the location information of the camera for each of the plurality of mutual areas to be detected, and providing the stored information to the operation server. The method of claim 8,
Wherein the mutual area includes a character area formed by a plate-shaped area or a character block,
Wherein the text information in the mutual region includes text or numbers. The method of claim 8,
Wherein,
Wherein the text information includes at least one of a foreign language character, a special symbol, and a number in the mutual area, or if the text information is not recognized from the mutual area, inputting or modifying the text information from the manager.

Images