JP2005100193A - Invasion monitoring device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an invasion monitoring device capable of monitoring presence/absence of an invader to a predetermined area with high reliability. <P>SOLUTION: In this invasion monitoring device 1 for monitoring presence/absence of the invader to the predetermined monitoring area, a binarization threshold value, which is used when information about a moving object is fetched by processing visible image data D1 of the monitoring area obtained by a first camera 1, is set (steps 34 and 35) according to a determination result (a step 33) of presence/absence of the invader based on heat wave image data D2 of the monitoring area obtained by a second camera 3. Therefore, when the threshold value for binarization of the visible image data D1 of a necessary part is lowered only if invasion is detected according to the heat wave image data D2, sensitiveness of invader detection based on the visible image is increased for surely detecting the invader according to the visible image. In this way, reliability of a monitoring result can be improved greatly. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to an intrusion monitoring apparatus suitable for monitoring the presence or absence of an intruder in a predetermined area.

  Conventionally, in order to monitor the presence or absence of an intruder in a room or the like, for example, an intrusion monitoring apparatus using visible image data from a monitoring camera installed in an appropriate place indoors or outdoors is used. This type of conventional intrusion monitoring apparatus detects the moving object in the visible image by processing the obtained visible image data using an analysis processing program, and detects the moving object based on the detected movement and size of the moving object. Is configured to determine whether or not it is an intruder.

  The conventional intrusion monitoring device using visible image data is caused by misinformation or masking due to wearing infrared heat shield clothing, etc., as in the case of a device configured to detect and monitor infrared rays emitted from intruders. There is no problem of not being aware of the unmonitorable state. However, there are many cases where intruders cannot be accurately detected due to optical noise generated from various environments, such as intruders cannot be distinguished from small animals, and it is difficult to perform highly reliable monitoring. Have.

  The objective of this invention is providing the intrusion monitoring apparatus which can solve the problem in the said prior art.

  Another object of the present invention is to provide an intrusion monitoring apparatus capable of reliably monitoring the presence or absence of an intruder in a predetermined area.

  In order to solve the above problems, the present invention focuses on infrared rays such as far infrared rays emitted from intruders, and includes a first camera for obtaining visible image data in a predetermined monitoring area, and infrared rays in the monitoring area. A second camera for obtaining image data, and combining information on the size and position of the intruder obtained by processing the infrared image data from the second camera with the processing algorithm for the visible image data Therefore, a highly reliable intruder monitoring function is realized. The infrared image data may be, for example, far-infrared image data using far infrared rays. If far-infrared rays are used, a monitoring image can be obtained even in a place with low illuminance, and a living body can be detected.

  According to the invention of claim 1, in an intrusion monitoring apparatus for monitoring the presence or absence of an intruder in a predetermined monitoring area, a first camera for obtaining visible image data of the monitoring area, and an infrared image of the monitoring area A second camera for obtaining data, a first determining means for processing the infrared image data to determine whether there is an intruder in the monitored area, and processing the visible image data Data processing means for extracting information on a moving object; setting means for setting a binarization threshold value of the visible image data in response to the determination result of the first determination means in response to the first determination means; and the data Binarizing the processing data from the processing means using the threshold value set by the setting means, and a second determination means for determining whether or not an intruder exists in the monitoring area; Preparation Intrusion monitoring device is proposed which is characterized in that the.

  According to the present invention, since the intruder is monitored based on the visible image data and the infrared image data of the monitoring area, it can cope with various intrusion conditions of the intruder and perform highly reliable intrusion monitoring. be able to.

  FIG. 1 is a schematic configuration diagram showing an embodiment of an intrusion monitoring apparatus according to the present invention. The intrusion monitoring device 1 is a device for monitoring the presence or absence of an intruder in a predetermined monitoring area such as a store, an office room, or a factory, and includes a first camera 2 and a second camera 3.

  The first camera 2 is a camera for obtaining visible image data in a predetermined monitoring area, and is a camera device for imaging a visible image having a known configuration including a lens 2A and a CCD element 2B. On the other hand, the second camera 3 is a camera for imaging an infrared image in a predetermined monitoring area. In the present embodiment, the second camera 3 is configured as a far-infrared camera device that captures far-infrared rays emitted from an intruder and obtains a heat-ray image, and includes a lens 3A and a heat-ray sensor array 3B. ing. If far-infrared rays are used, a monitoring image can be obtained even in a place with low illuminance, and a living body can be detected. However, the 2nd camera 3 is not limited to this, It can also be comprised as a camera for obtaining the infrared image by infrared rays other than a far infrared ray.

  FIG. 2 is a diagram showing the configuration of the second camera 3, and heat rays (far infrared rays) from the monitoring area are incident on the hemispherical reflecting mirror 3C through the lens 3A, and the light reflected by the reflecting mirror 3C is reflected in the heat ray sensor array. It is configured to be incident on 3B.

  Returning to FIG. 1, the visible image signal S <b> 1 obtained by the first camera 2 and the heat ray (far infrared) image signal S <b> 2 obtained by the second camera 3 are sent to the analog / digital (A / D) converter 4. Here, the image data is converted into corresponding digital data, and visible image data D1 and heat ray (far infrared) image data D2 are obtained. Here, instead of the analog-to-digital conversion configuration by the analog / digital (A / D) converter 4, the visible image signal S1 and the heat ray (far infrared) image signal S2 are binarized using a simple comparison circuit. It is also possible to use circuit means designed to do so.

  The visible image data D1 and the heat ray image data D2 are sent to the image processing unit 5, and the image processing unit 5 performs a sensing process according to a predetermined algorithm for combining the visible image data D1 and the heat ray image data D2. When it is determined that there is an intruder by the data processing in the image processing unit 5, the visible image data D1 and the heat ray image data D2 are sent to the image compression unit 6, and the image compression unit 6 displays the visible image data D1 and the heat ray image data. D2 is compressed and processed, and both compressed image data DX are sent to the communication unit 7. A general-purpose method such as J-PEG and an original method can be used as an image compression algorithm in the image compression unit 6.

  In response to the input of both compressed image data DX, the communication unit 7 transmits both compressed image data DX to a predetermined monitoring center. As a communication path between the communication unit 7 and the monitoring center, a general public line, a dedicated line, and other appropriate communication paths can be used.

  An alarm generation unit 9 is connected to the image processing unit 5, and when the image processing unit 5 determines that there is an intruder as in the conventional case, the image processing unit 5 determines the intrusion to the alarm generation unit 9. The signal J is output, and the alarm generation unit 9 sends the intrusion alarm signal K to the communication unit 7 in response to the intrusion determination signal J. In response to the input of the intrusion alarm signal K, the communication unit 7 sends the intrusion alarm signal K to a monitoring center (not shown) as in the case of processing both compressed image data DX.

  On the other hand, the visible image data D1 is always recorded in the silicon memory unit 8 in a compressed format for the latest several frames. When it is determined that there is an intruder in the image processing unit 5, an intrusion determination signal J is input to the silicon memory unit 8, and visible image data before and after the determination time can be output.

  In the I / O unit, an output signal from an external security sensor installed around the camera can be captured and used as a trigger signal for capturing an image. At this time, in addition to the ON / OFF state of the sensor, image data before and after can be output.

  The image processing unit 5 in FIG. 1 is configured using a microcomputer, and has a configuration that performs an intrusion determination function based on an image by causing the microcomputer to execute an image data processing program. Here, the image processing unit 5 processes the heat ray image data D2 to determine whether there is an intruder in the monitored area, and the visible image data D1 to process the moving object. A data processing function for retrieving the information, a setting function for setting a binarization threshold value of the visible image data D1 according to a determination result obtained by the first determination function, and a processing data obtained by the data processing function A binarization process is performed using a threshold set by the function, and a second determination function for determining whether or not an intruder exists in the monitoring area is provided.

  Next, the image processing unit 5 will be described in detail with reference to FIGS. 3 and 4.

  FIG. 3 is a flowchart showing a heat ray image data processing program 20 for data processing for intruder detection based on the heat ray image data D2. In step 21, (1) binarization of the heat ray image, (2) labeling, and (3) circumscribed rectangular shape processing are executed on the input heat ray image data D2, and the intruding object is detected from the heat ray image. Extract information. Based on the extracted information thus obtained, the distance Z to the object based on the extracted information is obtained in step 22, and the height (height) H of the object based on the extracted information is obtained in step 23. In step 24, the position B of the object and the width X of the object are obtained based on the extracted information.

  Thereafter, in step 25, it is determined whether or not the height (H) and width (X) of the object based on the extracted information exceed respective predetermined threshold values. If it is determined that both the height and width exceed the respective threshold values, the extracted information is determined to be intruder information, the determination result in step 25 is YES, and the intruder detection flag F is set in step 26. 1, the execution of the heat ray image data processing program 20 is completed. On the other hand, if it is determined that the height and width do not exceed the threshold values, the determination result in step 25 is NO, the intruder detection flag F is set to 0 in step 27, and the execution of the heat ray image data processing program 20 is executed. finish.

  FIG. 4 is a flowchart showing the visible image data processing program 30 for processing the visible image data D1. When the execution of the visible image data processing program 30 is started, in step 31, (1) background image update processing for the visible image data D1 from the analog / digital (A / D) converter 4, and (2) Background difference image generation processing is performed. The process of step 31 is a constant process that is always executed, whereby the intruding object image data is extracted from the visible image data D1.

  In the next step 32, the intrusion object image data obtained in step 31 is subjected to the same data processing as that in steps 21 to 24 in FIG. 3, and the first camera 2 captures the intrusion object image data based on the intrusion object image data. The distance Z, height H, position B, and width X are determined for the intruding object.

  In the next step 33, it is determined whether or not the intruder detection flag F obtained by executing the heat ray image data processing program 20 is 1.

  If F = 1 in step 33, the determination result in step 33 is YES, and step 34 is entered, where the threshold level for binarization processing of the visible image data D1 is lowered and the detection sensitivity is increased. Processing is executed. On the other hand, if F = 0 in step 33, the determination result in step 33 is NO, and step 35 is entered. In step 35, processing for increasing the threshold level for binarizing the visible image data D1 and decreasing the detection sensitivity is executed.

  In steps 34 and 35 constituting the mutual interference function unit, the detection sensitivity setting for the process of extracting the information on the intruding object from the visible image data D1 is set as to whether or not the heat ray detection is performed by the heat ray image data D2. The detection sensitivity setting function is changed according to (whether or not the determination result in step 25 is YES).

  When the detection sensitivity is set high by executing step 34, the process enters step 36, where an image indicating an intruder is extracted by the binarization processing of the visible image data D1 (whether the visible image is detected). ) Is determined. If it is determined in step 36 that the visible image is detected, the determination result in step 36 is YES, and it is determined that there is an intruder in the given monitoring area. In step 37, the monitoring result that there is an intruder is output and visible. The execution of the image data processing program 30 is terminated.

  On the other hand, if it is determined in step 36 that no visible image is detected, the determination result in step 36 is NO, and step 38 is entered. In this case, since it is determined that there is no intruder based on the visible image data D1 although it is determined that there is an intruder based on the heat ray image data D2, in step 38, the heat ray is determined. The details of the heat ray information are analyzed based on the image data D2. This detailed analysis is an analysis for determining whether or not motion is detected in the heat ray image. Also by this detailed analysis, if the determination that there is an intruder is still valid, it is determined in step 39 that either the temperature difference or the extremely dark intrusion occurs. If it is determined that the intruder is extremely dark, it is determined that there is an intruder in the given monitoring area. In step 37, a monitoring result indicating that there is an intruder is output to display the visible image data processing program 30. The execution of is terminated.

  Next, processing when the determination result of step 33 is NO and the process enters step 35 will be described. After the detection sensitivity is set low in step 35, in step 40, it is determined whether or not an image indicating an intruder has been extracted by the binarization processing of the visible image data D1 (whether or not the visible image is detected). The If it is determined in step 40 that a visible image is detected, the determination result in step 40 is YES, and it is determined that there is an intruder in the given monitoring area. In step 41, the monitoring result that there is an intruder is output and visible. The execution of the image data processing program 30 is terminated.

  On the other hand, if it is determined in step 40 that no visible image is detected, the result is that the image of the intruding object is not extracted based on either the visible image data D1 or the heat ray image data D2. The determination result is NO, and the process returns to step 31.

  If it is determined in step 40 that there is a visible image detection, a heat ray image of the intruder could not be obtained even though a detection result indicating that there is an intruder has been obtained based on the visible image data D1. Step 41 is entered, where it is determined that the intrusion is wearing the heat shield clothing or the masking of the second camera 3, and the monitoring result that there is an intruder is output.

  According to the image processing unit 5, as described above, the intruding object is detected by using the visible image data D1 by the first camera 2 and the heat ray image data D2 by the second camera 2, respectively. By comprehensively considering the results, it is possible to monitor intruders in extremely dark conditions, monitor intruders by wearing heat-blocking clothing, as well as grasping the presence or absence of work on the surveillance camera. Intruders can be monitored more reliably than in comparison.

  Also, only when there is intrusion detection using the heat ray image data D2, the sensitivity for intruder detection based on the visible image is increased by lowering the threshold value for the binarization processing of the visible image data D1, so that the visible image data D1 is visible. Since intruder detection is reliably performed using images, it is possible to reliably detect mere intruders as well as to detect intrusions by wearing heat-blocking clothing. Thus, the reliability of the monitoring result can be greatly improved.

1 is a schematic configuration diagram showing an embodiment of an intrusion monitoring apparatus according to the present invention. The figure which shows the structure of the 2nd camera 3 shown in FIG. The flowchart which shows a heat ray image data processing program. The flowchart which shows a visible image data processing program.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Intrusion monitoring apparatus 2 1st camera 3 2nd camera 4 Analog / digital (A / D) converter 5 Image processing part 6 Image compression part 7 Communication part 8 Silicon memory part 9 Alarm generation part B Position D1 Visible image data D2 Heat ray (Far infrared) image data DX both compressed image data H height J intrusion detection signal K intrusion alarm signal S1 visible image signal S2 heat ray (far infrared) image signal X width Z distance

Claims (1)

In an intrusion monitoring device for monitoring the presence or absence of an intruder in a predetermined monitoring area,
A first camera for obtaining visible image data of the surveillance area;
A second camera for obtaining infrared image data of the surveillance area;
First determination means for processing the infrared image data to determine whether an intruder exists in the monitoring area;
Data processing means for processing the visible image data and extracting information on the moving object;
In response to the first determination means, a setting means for setting a binarization threshold value of the visible image data according to the determination result of the first determination means;
Second discrimination means for binarizing the processing data from the data processing means using the threshold value set by the setting means to determine whether or not an intruder exists in the monitoring area And an intrusion monitoring device.

Images