KR100853744B1 - Image supervision unit and the method for detect circumstance change - Google Patents

Abstract

An image monitoring unit and a method for detecting a change of circumstances are provided to detect a change of circumstances in a monitoring area and report the detected change directly to monitoring people, thereby minimizing the determination error of the monitoring persons and the number of monitoring people. A monitor image setting unit(110) sets at least one reference image in a monitoring area. A compared area setting unit(120) sets at least one compared area in the reference image. An event setting unit sets one of a dynamic motion event and a static motion event and sets the dynamic motion event and the static motion event differently to each compared area when plural compared areas are set to at least one reference image. An image comparing unit compares an image part corresponding to a compared area in the reference image with an image part corresponding to a compared area in a current image and generates an event signal when motion set to the event setting unit is generated in at least one compared area. The comparison of the compared area is performed by using one of one-to-one comparison in pixel data, image comparison by frame, and motion vector comparison.

Description

Image supervision unit and the method for detect circumstance change}

1 is a block diagram of an image sensing apparatus according to the present invention.

2 is a signal flow diagram for a situation change detection method according to the present invention.

3 shows an embodiment of a situation change detection method according to the present invention.

4 shows another embodiment of a situation change detection method according to the present invention.

5 shows another embodiment of a situation change detection method according to the present invention.

The present invention relates to an image surveillance apparatus, and more particularly, to an image sensing apparatus capable of minimizing a mistake of a monitor operator and the number of monitor personnel by detecting a change in a situation of a monitoring place and generating an appropriate event.

Closed Circuit Television (CCTV) is a television system installed in a restricted area. It has a characteristic that a camera is connected to a television monitor and a signal does not propagate into the air. CCTVs were commonly used by public institutions for crackdowns on traffic violations, crime prevention, and use, but are now widely used in homes for security purposes.

In crowded airports, patrol guards patrol all areas at all times, although security guards patrol at regular intervals. Therefore, CCTVs will be installed in places where surveillance through patrols can be neglected, and the situation will be monitored to monitor crimes. In areas with large areas to be monitored, such as airports, a significant number of CCTVs must be installed. Since the control station that monitors the situation information on the monitoring site transmitted from CCTV cannot be enlarged indefinitely, one monitor displays the signals transmitted from multiple CCTVs. It is common to allow monitoring. However, this still requires a large number of monitors to monitor the status information of the monitoring site transmitted from CCTV.

In the exhibition halls or museums where expensive objects are displayed, such as works of art, CCTVs are monitored in addition to regular patrols by security guards to prevent theft of expensive objects on display. CCTVs are needed even in department stores. For example, in order to prevent theft, in addition to being installed in a general store, it is generally installed in an elevator or the like. In recent years, escalator accidents are frequent, and these accidents occur mainly in young children, so safety personnel regularly patrol or stay to monitor electrical failures such as escalator stop lights.

Modern factories tend to operate with mechanical automation systems instead of borrowing human hands. In the case of a factory operated by an automated system, CCTVs are installed in places where machine malfunctions occur and monitored so that problems can be dealt with immediately. Even where the factory is not automated, managers who manage the entire process often use CCTV to monitor that certain processes are stuck or delayed.

Although CCTV is used in various places as described above, it is the same that the monitor agent must directly see and judge the situation of the place to be monitored through the monitor. Therefore, the more CCTV installations, the greater the number of monitor personnel. However, the number of monitors is not only directly related to the cost of the place, but the human eye's judgment of the situation always has room for error.

An object of the present invention is to provide an image sensing apparatus capable of minimizing a mistake of a monitor operator and the number of monitor personnel by detecting a change in a situation of a monitoring place and generating an event corresponding thereto.

Another technical problem to be achieved by the present invention is to provide a situation change detection method for detecting a situation change for a monitoring place and generating an event corresponding thereto.

The video surveillance apparatus according to the present invention for achieving the above technical problem includes a surveillance image setting unit, a comparison area setting unit, an event setting unit and an image comparing unit. The surveillance image setting unit sets at least one reference image for the surveillance zone. The comparison area setting unit sets at least one comparison area in the reference image. The event setting unit sets a dynamic motion event and a static motion event. The image comparator compares at least one current image with the set comparison area and generates an event signal and a control signal corresponding to the comparison result.

The situation change detection method according to the present invention for achieving the above another technical problem comprises an initial condition setting step, image matching step and motion comparison step. The initial condition setting step sets a reference image, a comparison area, and an event generation condition in response to a control signal. The image matching step encodes the current image into an image corresponding to the comparison area. The motion comparison step compares the encoded current image with the comparison area and generates an event signal and the control signal.

Hereinafter, specific embodiments of the present invention will be described in detail with reference to the accompanying drawings.

The core idea of the present invention is to determine a change in the surveillance zone when monitoring a certain zone in the video sensing device and to generate an appropriate event so that the monitoring personnel can recognize it. To this end, the reference image information for a certain surveillance zone to be monitored is stored in advance, and the current image information received from the camera is compared with the stored reference image information, and then a certain event corresponding to the comparison result is generated. Will.

The reference image information includes image information selected from images input from a camera, in addition to image information forcibly set and stored by a user. In addition, the change of the monitoring zone is not only the change of the motion such as the movement of the objects included in the monitoring zone, the addition of new objects, but also the monitoring target, which must be continuously operated, such as the operation of the escalator. This includes changes in Non Motion.

The image information comparison used in the present invention includes distinguishing at least one comparison area for one image and comparing the current image information with reference image information set for each comparison area so as to detect a change in movement. In addition, the comparison of the image signals may be performed by comparing pixel data one-to-one or comparing images on a frame-by-frame basis, and using basic frame data and a motion vector associated with the basic frame data. It can be implemented by applying the method.

Here, the event refers to various forms of reporting and action means such as storage of an image in question, alarm, message, or mail.

1 is a block diagram of an image sensing apparatus according to the present invention.

Referring to FIG. 1, the image sensing apparatus 100 may include a surveillance image setting unit 110, a comparison area setting unit 120, an event setting unit 130, an image comparing unit 140, and an event generating unit 150. It is provided.

The surveillance image setting unit 110 may include at least one basic image (V1-1, V1-2 to V1-N, N is an integer) or at least one current image (V2) for a certain region (hereinafter, referred to as a surveillance region) to be monitored. -1, V2-3, -V2-N and the base image (V1-1, V1-2-V1-N) or the current image (V2-1, V2) in response to the control signal CON -3, ~ V2-N) is used to set the reference image (Ref-1, Ref-2 ~ Ref-N). The reference image may be set to the base image (V1-1, V1-2 to V1-N) or the entire current image (V2-1, V2-3, to V2-N) as a reference image or the base image (V1). -1, V1-2 to V1-N) or at least a certain range of surveillance zones selected from the entire surveillance zones included in the current image (V2-1, V2-3, V2-N) are selected. One reference image (Ref-1, Ref-2 to Ref-N) can be set.

Here, each of the basic images V1-1, V1-2, and V1-N may be images corresponding to a plurality of surveillance zones, or may be an image obtained by dividing a plurality of images of one surveillance zone into a plurality of images. The current images V2-1, V2-3, and V2-N may also be set in the same manner as the basic images V1-1, V1-2, and V1-N.

In the state in which the control signal CON generated by the image comparator 140 is disabled, the reference images Ref1, Ref2, and RefN are generated using the base images V1-1, V1-2, and V1-N. ). The control signal CON is enabled when the situation of the monitoring area is not changed for a certain time. In this case, the reference image may be set using the current image, and the reference image may be replaced with the existing reference image, or the reference image may be added as the reference image without changing the reference image.

The basic image (V1-1, V1-2 to V1-N) and the current image (V2-1, V2-2 to V2-N) are generally generated from an image device such as a camera and converted into a digital signal. It is a signal, and the basic image is an image previously photographed for the surveillance zone, and the current image refers to an image currently input through the camera.

The comparison area setting unit 120 may compare the entire area or at least one area indicated by the at least one reference image Ref1, Ref2 to RefN set by the surveillance image setting unit 110 (Ref-Area1, Ref). -Area2 ~ Ref-AreaN) are set.

Although the entire reference image may be set as a comparison area as it is, when it is desired to monitor the change of dynamic movement or the change of static movement only in a part of the area indicated by the reference image, the comparison area may be set for a portion of the reference image. If there are two or more places to be precisely monitored in the image information of one surveillance zone, two or more regions of one reference image may be set as a comparison region.

The reference images (Ref1, Ref2 to RefN) set by the surveillance image setting unit 110 have information about the entire constant 3D space, but all of the image information about the captured constant overall space may be needed, Sometimes only the image information of a certain space is needed. In addition, the monitoring of the entire space is necessary, but since the entire space is so extensive, there may be a problem that the monitoring may not be completed when the image information of the entire space is set as the reference image (Ref1, Ref2 to RefN). It monitors this by dividing the information about a certain total space photographed into a plurality of parts.

In summary, the comparison area setting unit 120 may set the comparison areas Ref-Area1, Ref-Area2 to Ref-AreaN under the following two conditions.

1. Image information about the whole space as it is

2. Select only certain parts of the entire video information

Since only the image information of the space required by the surveillance image setting unit 110 is set as the reference image, the comparison area setting unit 120 may perform more precise discrimination. That is, even in the case of the reference image set to the required space in the surveillance image setting unit 110 may include image information that is unnecessarily necessary in the setting technology, the comparison region setting unit 120 for more precise comparison even in such a case There is a feature that can be reselected.

The event setting unit 130 sets an event occurrence condition for the comparison area set by the comparison area setting unit 120 and generates an event setting signal E-Set corresponding thereto. For example, by comparing the comparison area of the reference image and the comparison area of the input current image, it can be set to generate an event when a motion of a predetermined level or more is detected or to generate an event when a motion is not detected. In addition, when two or more comparison areas (Ref-Area1, Area2 to AreaN) are set in the comparison area setting unit 120 with respect to one reference image, the event setting unit 120 may generate different event generation conditions for each comparison area. Can be set. For example, when the three comparison areas Ref-Area1, Area2, and Area3 are set in the comparison area setting unit 120, the first comparison area Ref-Area1 and the third comparison area Ref-Area3 are dynamic. Motion detection may be set as an event generation condition, and non-motion detection may be set as an event generation condition for the second comparison area Ref-Area2.

In response to the event setting signal (E-Set), the image comparing unit 140 includes the reference image set by the surveillance image setting unit 110 and at least one current image V2-1, V2-2 to V2-N. ), An image extractor 141, a comparison area matcher 142, and a motion comparator 143.

The image extractor 141 extracts an image of the comparison region from the reference image, and extracts an image of the comparison region among the current images continuously input. When a plurality of comparison areas is set, each image corresponding to the comparison area is extracted from the reference image and the current image. The comparison region matching unit 142 encodes the current image and matches the reference image so that the comparison region of the extracted current image and the comparison region of the reference image can be compared one-to-one. The motion comparison unit 143 compares the encoded current image with the reference image to determine a state of control signals CON1, CON2, and CONN, and determines an event signal ES1 according to an event setting signal E-Set. , ES2, ~ ESN). Motion comparison uses a method of comparing pixel data in a one-to-one or frame-wise image. Also, basic frame data and a motion vector associated with the basic frame data may be used.

The image comparison unit 140 compares the image portion corresponding to the comparison region set by the comparison region setting unit 120 with the image portion corresponding to the comparison region among the current images. In addition, when there are a plurality of comparison areas set by the comparison area setting unit 120, the image part corresponding to the comparison area among the reference images and the current image part corresponding thereto are compared.

Current image (V2-1, V2-2 ~ V2-N) transmitted to the image comparison unit 140 and the base image (V1-1, V1-2 ~ V1-N) transmitted to the surveillance image setting unit 110 This may be from the same camera, but it does not matter. However, it is assumed that the image information generated from the same camera for convenience of description.

If the image comparison unit 140 compares the comparison region of the reference image with the comparison region of the current image and determines that the current image is different from the reference image, that is, the event occurrence condition set by the event setting unit 130 (E-Set). ), The event signal ES corresponding thereto is outputted. Here, as a result of comparing the comparison region of the reference image and the comparison region of the current image, the difference between the current image and the reference image corresponds to the case of detecting not only the dynamic movement but also the static movement. The event signals ES for the plurality of comparison areas to be compared are each generated independently. The event signal ES is a signal instructing to generate an event because a problem occurs in the surveillance space.

In addition, if it is determined that the current video and the reference video match within a certain tolerance range, the continuous time determined to match is set as the invariant time (Ts), and the invariant time (Ts) is set in advance. Compared to the time Tth, the control signal CON can be enabled when the invariant time Ts is equal to or greater than the threshold time Tth. The threshold time Tth is any time determined according to the situation of the surveillance zone.

The event generator 150 generates a warning signal Event in response to the event signal ES. For example, the event sound emitter emits a warning sound through a speaker. In addition, the present invention may be implemented in various ways, such as recording the current image V2, a warning sound or a warning message sent to the mobile communicator, or an e-mail transmitted to the administrator.

When displaying a plurality of monitoring zones through a plurality of monitors, a warning signal (Event) can be generated by blinking a monitor corresponding to a zone where an event signal (ES) is generated or by operating a warning light connected to the monitor. have. When displaying a plurality of monitoring zones on a single monitor, only an area where the event signal (ES) occurs among the plurality of monitoring zones blinks or a warning number assigned to the corresponding zone is alerted through a speaker. Becomes

As described above, the image monitoring device according to the present invention detects not only the change of the motion but also the change of the non-motion. Hereinafter, the monitoring of the airport and the monitoring of the escalator respectively change and change the motion of the motion. It demonstrates as an example of monitoring a change of movement.

The image sensing apparatus according to the present invention first stores image information (reference image) for the surveillance zone. The camera continuously receives an image of the surveillance zone and compares it with the reference image. If someone induces an object such as an explosive in the surveillance zone, the image comparator 140 of the image sensing device can immediately detect this and send an alarm signal to the monitor or control system. do. The problem can be solved through the above steps.

When you want to monitor the operation of the escalator in the department store, first store the video information (reference video) for the monitoring space. By comparing this with the current video signal, it is possible to determine whether the escalator is stationary or in operation.

In addition, after storing the image information (reference image) when the problem does not occur in the surveillance space in the factory, and compare it with the image signal of the current surveillance space, it can be monitored whether the problem occurred.

2 is a signal flow diagram illustrating a situation change detection method according to the present invention, and FIGS. 3 to 5 are views illustrating an embodiment of the situation change detection method according to the present invention.

Referring to FIG. 2, the situation change detection method 200 includes an initial condition setting step 210, an image matching step 230, a current image and a reference image comparison step 250, and an event generation step 270. .

The initial condition setting step 210 includes a reference image setting step 211, a comparison area setting step 212, and an event generating condition setting step 213.

In the reference image setting step 211, at least one reference image corresponding to the reference image is set using at least one base image or the current image of the surveillance zone to be compared based on the control signal CON.

In the comparison area setting step 212, the entire area or the partial area indicated by the reference image is set as the comparison area. The comparison area may be set to one area as shown in FIGS. 3 and 4, and may be set to a plurality of areas as shown in FIG. 5.

The event occurrence condition setting step 213 sets an event occurrence condition. Here, the event occurrence condition refers to the case of detecting dynamic movement or static movement in the monitoring area. In addition, the plurality of comparison regions may be set to generate an event. For example, FIG. 3 is set to generate an event when a dynamic movement is detected in the comparison area, that is, when a motion of a picture occurs, and FIG. 4 does not generate a movement of an escalator when a static motion is detected within the comparison area. If not, the face event can be set to fire. In addition, FIG. 5 illustrates that an event occurs when a movement of an object occurs in the first comparison area Ref-Area1, and an event occurs when the movement of the conveyor does not occur in the second comparison area Ref-Area2. Can be set to occur.

The image matching step 230 includes an image extraction step 231 and a matching step 232. The image extraction step 231 extracts an image of the comparison region from the current image which is continuously input to the image of the comparison region among the reference image. In the matching step 232, the current image is encoded and matched with the reference image so that the predetermined region of the extracted image and the comparison region of the reference image can be compared one-to-one.

The motion comparison step 250 includes a motion detection step 251, a variable time calculation step 252, and a time comparison step 253.

The motion detection step 251 detects whether there is a dynamic movement or a static movement in the surveillance zone. In other words, it is determined whether the reference image and the current image represent the same image, that is, the same situation. In this case, when comparing two images one-to-one, there may be a certain error between the images to be compared, so that the two images may be determined to be identical if they are within a preset tolerance range. This tolerance is a value that can be arbitrarily changed depending on the system, the designer's experience, and how it is applied to the actual situation. If the result of the comparison determines that it is out of the tolerance range (Yes), an event signal (ES) is generated, and if it is determined to be within the tolerance range (No), the current image encoding step 230 is continued.

For example, as shown in FIG. 3, when the event generation condition is the detection of the dynamic movement, when the reference image and the current image are the same image or the same situation, the dynamic movement is not detected, and when the event is not the same, the dynamic movement is detected. . On the contrary, as shown in FIG. 4, when the event occurrence condition is the detection of the static movement, when the reference image and the current image are the same image, the static movement is detected and when the event is not the same, the static movement is not detected.

In addition, as shown in FIG. 5, the plurality of comparison areas and the current image of the reference image to be compared generate respective events independently of each other according to the event occurrence condition.

In the non-variable time calculation step 252, when it is determined that the current image and the reference image are the same (No) as a result of the comparison in the comparing step 251, the time during which the situation of the surveillance area is not changed. Accumulate phosphorus fluctuation time (Ts). Here, Ts is different from the initially set reference image due to the change of the situation of the surveillance zone, except that the situation of the surveillance zone is the same as that of the initially set reference image. It also includes the time you hold.

The time comparison step 253 compares the set threshold time Tth with the calculated no-variation time Ts, and any subsequent means when the no-variable time Ts is smaller than the set threshold time Tth (No). However, the control signal CON is enabled when the non-variable time Ts is greater than the set threshold time Tth (Yes). Here, the threshold time Tth is an arbitrary time determined according to the situation of the monitoring zone, and may be set in the initial condition setting step 210.

In the initial condition setting step 210, when the control signal CON is disabled, the base image is generated using the base image. However, when the control signal CON is enabled, the current image is used. To generate a reference image corresponding thereto.

The event generation step 270 generates a warning signal Event in response to the event signal ES. Here, the event may be implemented in various ways, such as generating a warning sound through a speaker, recording a current image V2, a warning sound or a warning message sent to a mobile communicator, or an e-mail transmitted to an administrator. The warning signal is generated independently for each image to be compared, thereby enabling collective control of a surveillance area in which a plurality of cameras are installed.

In the above description, the technical idea of the present invention has been described with the accompanying drawings, which illustrate exemplary embodiments of the present invention by way of example and do not limit the present invention. In addition, it is obvious that any person having ordinary knowledge in the technical field to which the present invention belongs can make various modifications and imitations without departing from the scope of the technical idea of the present invention.

As described above, the image sensing apparatus and the situation change detection method used in the image sensing apparatus according to the present invention detect the situation change of the monitoring place and report the situation directly, thereby determining the mistake of the monitoring agent and the number of the monitoring personnel. There is an advantage that can be minimized.

Claims (17)

A surveillance image setting unit for setting at least one reference image for the surveillance region; A comparison area setting unit for setting at least one comparison area in the reference image; Any one of a dynamic motion event for generating an event when a motion occurs in the at least one comparison area and a static motion event for generating an event when no motion occurs in the at least one comparison area may be set. An event setting unit configured to differently set the dynamic motion event and the static motion event in each comparison area when a plurality of comparison areas are set in the comparison area setting unit with respect to one of the reference images; And The image portion corresponding to the comparison region of the reference image and the image portion corresponding to the comparison region among the current images input after setting the reference image are compared. When the comparison region is set in plural, the image corresponding to the comparison region among the reference images is compared. And an image comparator configured to compare an image part and an image part corresponding to a comparison area among the current images corresponding thereto, and generate an event signal when a movement set by the event setting part occurs in at least one comparison area. The comparison area is compared by using a comparison method of any one of one-to-one comparison of pixel data, image comparison on a frame basis, and motion vector comparison. delete The method of claim 1, wherein the comparison area, And an entire region or at least one divided region of the reference image. The method of claim 1, wherein the image comparator, An image extracting unit which extracts at least one image from among current images continuously input; A comparison area matching unit matching the extracted comparison area of the current image with the comparison area of the reference image; And And a motion comparator configured to compare the matched comparison area and generate an event signal when at least one of a dynamic motion and a static motion occurs. The method of claim 4, wherein If it is determined that the comparison region portion of the extracted current image and the comparison region portion of the reference image are the same as the result of the comparison in the motion comparison unit, the non-variable time accumulated in the time determined to be the same and the preset limit And comparing a time and generating a control signal when the invariant time is equal to or larger than the limit time. The method of claim 5, wherein the surveillance image setting unit, And setting the reference image using the at least one base image, and setting the reference image using the at least one current image when the control signal is enabled. delete delete delete The method of claim 1, And an event generator for implementing one action of recording the current video, generating an alarm sound, sending a warning message, and warning mail in response to the event signal. delete delete delete delete delete delete delete

Images