JP4650346B2 - Surveillance camera device - Google Patents

Description

  The present invention relates to a monitoring camera device with a sensor added to an interphone device or the like.

  Due to recent improvements in crime prevention awareness, convenience such as being able to confirm the face of a visitor or recording the face of a visitor when absent is evaluated, and a door phone camera system for residential use has become widespread. In addition, with the widespread use, door phone camera systems have been provided in which surveillance cameras are added as terminals for the purpose of preventing intrusion into houses, and the system is improved so that the surrounding area can be warned.

  FIG. 4 shows an example of the schematic diagram. In this illustrated example, an interphone between a doorphone with a camera 100 installed at a gate G of a house H and a doorphone master 101 installed in the house H is used as an interphone. In a doorphone system in which a visitor captured by a camera of the doorphone slave unit 100 with a call is confirmed by an image displayed on the monitor device M of the doorphone master unit 101 and the image can be recorded, for example, on the outer wall of a house H When the surveillance camera device 102 with a human detection sensor is attached and the human detection sensor detects an intruder in the site 103, the detection information and an image captured by the imaging camera of the surveillance camera device 102 are sent to the door phone parent unit 101. The monitor device M displays the image, and the alarm device provided in the door phone master unit 101 issues the alarm while the monitor device M It is lit illumination device 104 is provided in the camera apparatus 102 is adapted to provide menacing intruder.

  As a human detection sensor provided in the monitoring camera apparatus 102, a human is detected from a heat ray sensor that detects a person by detecting a heat ray (infrared ray) emitted from a human body or an image captured by an imaging camera of the monitoring camera apparatus 102. There is an image sensor.

  By the way, since the image sensor can know the position of the intruder that is the detection target, for example, an area in which a tree that is easily misdetected as a moving object (intruder) is shaken by a wind or the like is planted. Detection area), and even if there is a tree that sways in the mask area due to wind, etc., the alarm is not issued, but changes in shadows and light are detected as moving objects (intruders) There is a weak point that will be emitted.

  On the other hand, although the heat ray sensor can monitor a plurality of detection areas with one pyroelectric element (heat ray detection element), it has a weak point that it is difficult to set a detection area for viewing a wide area. There is a feature that it does not react to the movement of the shadow.

As the above-described monitoring camera device 102, there is provided a device in which both sensors are provided side by side to compensate for the weak points of both sensors to reduce false detection (for example, Patent Document 1).
JP 2000-209573 (paragraph 0011, FIG. 1)

  However, even if the area where trees planted in the site 103 exist is set as a mask area, if the tree sways due to wind or the like and a branch protrudes from the mask area, the image sensor moves it. There is a problem that it is detected as an object.

  Moreover, in the case of a heat ray sensor, when the level of a heat ray beam changes due to fluctuations in a tree or the like, there is a problem of detecting this as a person.

  Therefore, even when human detection is performed using the detection information of both sensors, both sensors output detection information, and as a result, false alarms are made, and if these false detections continue continuously, the false alarms may continue. was there.

  The present invention has been made in view of the above points, and its purpose is a surveillance camera that reduces the risk of false detection of a person even if there is a movement of an object that is not to be detected that protrudes from the mask area. To provide an apparatus.

In order to achieve the above-described object, the invention of claim 1 has an imaging means for imaging a predetermined area, and a function for setting a mask area in an area of an image obtained by imaging by the imaging means, An image sensor having an image processing means for detecting a moving object from an image, determining whether the position of the object is within the set mask area and outputting detection information;
The image processing means, if the movement of an object that is a non-detection target in the setting of the mask area is equal to or greater than a certain number of times during a certain time, then the number of movements after a certain period of time is less than the certain number of times. Up to this point, the area of the mask region is enlarged so that the movement range of the object is within the mask region.

  According to the first aspect of the present invention, the mask area is enlarged by enlarging the area of the mask area when a non-detection target object such as a tree existing in the mask area moves and a part of the mask area protrudes or is likely to protrude. Since it can be accommodated in the area, there is little risk of erroneous detection of the object as a moving object (intruder).

  According to a second aspect of the present invention, in the first aspect of the present invention, the expansion direction of the mask region is a lateral direction with respect to the mask region.

  According to the second aspect of the present invention, it is possible to prevent erroneous detection due to a tree that sways laterally due to wind or the like without unnecessarily increasing the mask area.

  According to a third aspect of the present invention, in the first or second aspect of the present invention, the heat ray sensor outputs detection information from a change in the light receiving level of a heat ray beam from a detection region set in a region of an image obtained by imaging of the imaging means. And determination means for determining an alarm notification based on detection information from the image sensor and detection information from the heat ray sensor.

  According to the invention of claim 3, the detection performance can be further improved and the false alarm can be reduced by the combined use of the heat ray sensor.

  In the present invention, when a non-detection target object such as a tree existing in the mask area moves and the mask area is likely to protrude, the mask area is enlarged, so that the object is erroneously detected as a moving object (intruder). There is an effect that there is little fear.

Embodiments of the present invention will be described below.
(Embodiment 1)
FIG. 1 shows a configuration of a surveillance camera device according to the present embodiment. The surveillance camera device according to the present embodiment is attached to an outer wall of a house and is an imaging unit for imaging a predetermined area (for example, a site of a house). The imaging camera 1, the YUV / NTSC converter 2 that converts the YUV format image data output from the imaging camera 1 into an NTSC image signal, and the NTSC image signal output from the YUV / NTSC converter 2 From the imaging camera 1, a monitor system circuit including a transmission unit 3 that modulates with a predetermined modulation method and transmits the signal to a door phone parent unit (not shown) provided in the house via a signal line (not shown), for example. An image sensor 4 that detects a moving object (intruder) from the output image data, and an image area obtained by imaging by the imaging camera 1 is divided into a plurality of detection areas, and the respective detection areas are used. A heat detection sensor 5 detects the presence or absence of a person with a heat beam (infrared ray) and the detection information of each of the image sensor 4 and the heat ray sensor 5 to make a human detection determination. And a signal processing unit 7 that outputs a signal for generating sound or lighting a threatening illumination lamp (not shown) that irradiates the site.

  The imaging camera 1 uses a fixed imaging element such as a CCD or a CMOS image sensor as an imaging element for imaging a monitoring area in time series. A / D conversion is performed on a signal output from the imaging element to convert it into YUV data. It has a function to do.

  The image sensor 4 includes an image processing unit (image processing unit) 40 and a storage unit 41. The image processing unit 40 includes a contour extracting unit 40a, a moving contour extracting unit 40b, a moving region detecting unit 40c, and a region feature amount detecting unit. 40d and mask processing means 40e.

  Here, the contour extracting means 40a performs a differentiation process on the time-series grayscale image data output from the imaging camera 1 using a general SOBEL filter or the like to extract a contour image, and this extracted contour A process of storing the image data in the storage unit 41 for a certain period is performed.

  The moving contour extraction unit 40b performs logical synthesis using time-series contour image data, and performs a process of creating and extracting a composite image from only the contour in the moving state.

  The moving region extraction unit 40c extracts the region of the moving object by labeling the composite image obtained by the moving contour extraction unit 40. If the area to be extracted is represented by a circumscribed square area X as shown in FIG. 3, the amount of data is reduced and the calculation process is facilitated.

  The area feature quantity detection means 40d uses the contour image data stored in the storage unit 41 to extract the direction value of the moving contour portion to create a moving direction frequency distribution, and the created moving direction frequency distribution is characterized. As a quantity, it has a function of determining whether the area movement of the moving object is due to a disturbance or a person.

  By the way, the contours of humans have more curved parts than the linear components and have a complicated shape, so the edge directions are distributed in all directions, and the contours such as shadows of structures have many linear components and the frequency distribution in the moving direction. Tends to have a biased distribution.

  Therefore, a moving direction frequency distribution table for comparing and determining whether the moving area represents a person for the moving area is stored in the storage unit 41, for example, and the moving direction frequency distribution indicated by the table and the generated moving direction frequency are stored. By comparing the distribution, it is determined whether the region movement is caused by a person.

  The mask processing means 40e includes a movable object such as a gate in a predetermined area to be monitored (such as a residential premises) to be imaged by the imaging camera 1, and a tree that is moved by shaking of the wind. Is a means having a function of setting the mask area MA as shown in FIG. 3 so as not to detect it as a moving object (intruder), and also performs a mask area widening process as described later.

  In the image sensor 4 configured as described above, the region feature amount detection unit 40d of the image processing unit 4 outputs a result determined as a person to the signal processing unit 7 as detection information.

  On the other hand, the heat ray sensor 5 is provided together with the imaging camera 1, and the area (image area) captured by the imaging camera 1 is divided into a plurality of areas as indicated by the hatched portion in FIG. As a detection region, a heat ray optical unit 10 such as a lens that collects a heat ray beam (infrared beam) from these detection regions, a pyroelectric element 11 that receives the heat ray beam collected by the heat ray optical unit 10, and the pyroelectric element An amplifying circuit 12 that amplifies the detection signal of the element 11 and a comparator that compares the amplified output of the amplifying circuit 12 with a predetermined reference value and outputs detection information to the signal processing unit 7 when a change in the received light level is more than a predetermined value. 13.

  Note that the detection region shown in FIG. 2 is actually set to have a planar extension.

  The signal processing unit 7 functions as a determination unit for determining alarm notification when both the detection information of the image sensor 4 and the detection information of the heat ray sensor 5 are input, and the notification sound to the notification / alarm unit 40 based on the determination result. And a control unit that outputs a control signal for informing the site and a control signal for turning on the illumination lamp that illuminates the site, and sending each control signal to the notification / alarm unit 40 to provide the notification / alarm unit A warning sound indicating that there is an intruder in the householder is issued from the notification sound from 40, and the intruder is intimidated by the light of the illumination lamp.

  By the way, when a tree is planted in the site to be a monitoring area, if the tree is shaken by the wind, the image sensor 4 erroneously detects it as a moving object. Further, the heat ray sensor 5 also detects erroneously because the heat dose received due to the shaking of the tree changes, and the result signal processing unit 7 determines that there is an intruder based on the detection information of both sensors 4, 5, There is a risk of false alarms from the alarm unit 40.

  Therefore, in the present embodiment, the above-described mask processing means 40e removes an area where a moving object such as a tree or a gate is present from the detection area of the image sensor 4, that is, sets the mask area MA in the image processing unit 40 in advance. It has become. This setting may be performed by appropriate means. For example, as an initial setting, the mask area MA is set on the screen while the operator or the user confirms the area in the image obtained by the imaging camera 1 with the monitor device. Then, the setting data is sent to the image processing unit 40 through the signal processing unit 7 and passed to the mask processing unit 40e. Of course, the mask area MA may be set using other methods.

  In the setting example in the case of FIG. 3A, a large mask area MA is set in which the entire tree Tr can be accommodated even when the wind is not blown or the wind strength is below a predetermined value. This mask area MA is small. A plurality of small mask areas Ma are gathered to cover the entire tree Tr. The mask area MA 'at the right corner corresponds to an area where there is an object that moves with opening and closing, such as a gate.

  By the way, when the wind blows strongly, the tree Tr is greatly shaken to protrude from the mask area MA, and the image sensor 4 may detect the protruding part as a moving object due to the movement of the protruding part. Therefore, it is conceivable to set the mask area MA large from the beginning, but in this case, the detection area of the image sensor 4 is narrowed. Therefore, as shown in FIG. 3B, the tree Tr moves between the small mask areas Ma constituting the mask area MA, and the contour extracting means 40a, the moving contour extracting section 40b, the moving area detecting means 40c, the moving feature amount When detection is performed in the same manner as normal moving object detection by the detection means 40d, the built-in counter is counted up each time the detection is performed by the mask processing means 40e, and the number of movements counted is a certain number of times within a certain time. In this case, a process of additionally setting the small mask area Ma is performed so that the mask area MA is widened in the horizontal direction as shown in FIG. In the mask area MA, when the number of movements counted within a certain time after the widening process is less than a certain number, the process returns to the original width.

  Thus, in this embodiment, when the number of movements of the tree Tr in the mask area MA exceeds a predetermined number, the mask area MA is widened laterally by the mask processing means 40e, thereby increasing the movement of the tree Tr. However, the image processing unit 4 does not detect the movement of the tree Tr as a moving object (intruder), and the signal processing unit 7 does not detect the tree Tr. Even if the heat ray sensor 5 is mistakenly detected as a human by movement, the image sensor 4 does not falsely detect it. Therefore, it is not determined that there is an intruder, and the false alarm in the notification / alarm unit 6 can be prevented. Further, if the number of movements of the tree Tr within a certain period thereafter becomes less than the certain number, the mask area MA is returned to the original width, so that the detection area is not maintained narrow. There is little risk of missing an intruder.

  In the above-described embodiment, the number of movements of the tree Tr existing in the mask area MA is counted. However, before the above-mentioned count number is counted a predetermined number or more, the tree Tr is extracted from the mask area MA. Rather than immediately detecting the presence of an intruder even when a part of the area protrudes, the count of the number of movements within a certain period of time is continued, and the area of the mask area MA is increased when counting a predetermined number of times, The protruding part may be accommodated in the mask area MA.

It is a block diagram of one embodiment. It is explanatory drawing of the detection area | region of the heat ray sensor of one Embodiment. It is setting explanatory drawing of the mask area | region of one Embodiment. It is explanatory drawing of the intercom system provided with the monitoring camera apparatus.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Imaging camera 2 YUV / NTSC conversion part 3 Transmission part 4 Image sensor 40 Image processing part 40a Contour extraction part 40b Moving outline extraction means 40c Moving area extraction means 40d Area feature-value detection means 40e Mask processing means 41 Storage part 5 Heat ray sensor 50 Lens 51 Pyroelectric element 52 Amplifying circuit 53 Comparator 6 Notification / alarm unit 7 Signal processing unit

Claims (3)

An imaging means for imaging a predetermined area, and a function for setting a mask area within an area of an image obtained by imaging of the imaging means, and an object moving from the image is detected and the position of the object is set And at least an image sensor having image processing means for outputting detection information by determining whether or not the mask area is present,
The image processing means, if the movement of an object that is a non-detection target in the setting of the mask area is equal to or greater than a certain number of times during a certain time, then the number of movements after a certain period of time is less than the certain number of times. Until then, the area of the mask region is enlarged so that the movement range of the object is within the mask region. The surveillance camera device according to claim 1, wherein an enlargement direction of the mask region is a lateral direction with respect to the mask region. A heat ray sensor that outputs detection information from a change in the received light level of a heat ray beam from a detection region set in an image region obtained by imaging of the imaging means, detection information from the image sensor, and detection from the heat ray sensor 3. The surveillance camera device according to claim 1, further comprising a determination unit that determines an alarm notification based on the information.

Images