JP2004153445A - Monitor camera - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a monitor camera for taking automatically a clear photograph of both the foreground and the background of a person with different photographic conditions. <P>SOLUTION: A digital camera 11 repeats shooting at the interval of 5 seconds, and the taken picture is stored as a reference image in a reference image memory 21. When a detected signal is entered from a sensor 19 to a controller 16, the shutter of the digital camera 11 is released, and the taken picture is stored as a background image in a background memory 22. The background image and the reference image are compared with each other in an image comparing unit 23, and a variance image region is detected. A mask image is generated from the variance image, and the variance image region of the background image is covered. By controlling the digital camera 11, a focus/photometry area is set at the center of the mask image and the shooting is carried out, and the taken image is stored as a foreground image in a foreground image memory 24. A personal image is carved out from the foreground image in an image synthesizing unit 25 and composited in a mask image 34 of the background image 32. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a surveillance camera, and more particularly to a surveillance camera that can automatically and clearly shoot both images even when the foreground and background of a person or the like are significantly different.
[0002]
[Prior art]
As a surveillance camera, for example, a video camera installed so as to always photograph the vicinity of a cash register in a convenience store is known. When an incident such as robbery occurs, the video tape taken by the video camera can be played back to check the video. However, in order to create a wanted photo, a non-linear editing system captures an image of a criminal and converts it into digital image data. Many general consumer video cameras have about 400,000 Since pixels and a low resolution CCD are used, a clear image cannot be obtained.
[0003]
Therefore, it is conceivable to use a digital camera using a CCD with 1.5 to 3 million pixels as a surveillance camera even in the popular type, but the distance between the foreground of a person and the background and the background is large, and the brightness of each is also significantly different. In such a case, since the autofocus mechanism and the automatic exposure mechanism of the digital camera adjust the focus and exposure according to only one of the foreground and the background, there is a disadvantage that only one of them is clearly captured.
[0004]
To clearly capture both the foreground and the background with significantly different shooting conditions, use a slow shutter in accordance with a low-luminance background such as a night view, and operate the strobe device just before the shutter closes to bring the foreground to a person, etc. A camera that emits strobe light is known (see Patent Document 1).
[0005]
In addition, there is known a camera that can capture the brightness of two different images by multiple exposure (see Patent Document 2). This camera exposes a subject image as a base image on a silver salt film in the first shooting, and at the same time, an electronic image of the same subject is captured by an imaging device and displayed on a liquid crystal monitor. Next, only the imaging device is operated, and a new electronic image (composite image) is displayed over the base image.
[0006]
When the photographer changes the position of the camera, the position of the composite image with respect to the base image changes, so the composition is determined while viewing the display on the liquid crystal monitor. Further, since the brightness of the composite image changes by operating the brightness adjustment dial, the balance between the brightness of the base image and the composite image is adjusted while viewing the display on the liquid crystal monitor. The brightness adjustment dial is linked to the shutter control of the camera, and when the second shot is taken, a composite image with the balance of brightness between the composition desired by the photographer and the base image is exposed to the silver salt film. The
[0007]
[Patent Document 1]
JP-A-8-122844 [Patent Document 2]
JP 2000-321635 A
[Problems to be solved by the invention]
However, in the camera described in Patent Document 1, both the exposure with the slow shutter and the exposure with the strobe light are performed on the foreground person. Therefore, when the person moves during the exposure, the shooting with the slow shutter is performed. There is a drawback in that the image appears like a shadow of a photographed image by strobe light, resulting in a very unnatural photograph. In the camera described in Patent Document 2, both of the two images can be clearly captured, but the composition is determined and the brightness is adjusted by human operation (although the brightness adjustment can be automated). It is not suitable as a surveillance camera that assumes automatic shooting.
[0009]
It is an object of the present invention to provide a surveillance camera that can automatically and clearly capture both images even when the foreground and background of a person or the like are significantly different.
[0010]
[Means for Solving the Problems]
In order to achieve the above object, the surveillance camera of the present invention is installed at a predetermined position, and imaging means for photographing the monitored location at regular intervals, and detection means for detecting an intruder in the monitored location, When the detection means detects no intruder, the reference image storage means stores the image data of the photographed image taken by the photographing means as the image data of the reference image, and when the detection means detects the intruder First photographed image storage means for storing image data of a photographed image photographed by the photographing means as image data of the first photographed image, and an area for detecting a different image area that is different from the first photographed image compared with the reference image. A detection unit; a shooting control unit configured to control the shooting unit to perform shooting by aligning a focus position and a photometric area at a center of the different image region; and the shooting unit is focused at a center of the different image region. A second photographed image storage means for storing image data of a photographed image taken together with the position and photometry area as image data of the second photographed image, and a first image obtained by cutting out an image corresponding to the different image region from the second photographed image. It comprises image composition means for fitting and synthesizing the captured image or the reference image.
[0011]
In addition, an imaging unit that is installed at a predetermined position and images a monitored place at regular intervals, a reference image storage unit that stores image data of a reference image previously captured by the imaging unit, and the imaging unit And a region detecting means for detecting a different image region that is changed by comparing a captured image captured every fixed time with a reference image, and image data of the captured image in which the different image region is detected is stored in the first captured image. A first photographed image storage means for storing as image data, a photography control means for controlling the photography means and photographing with a focus position and a photometric area aligned with the center of the different image area, and the photography control means. Second photographed image storage means for storing image data of the photographed photographed image as image data of the second photographed image; and an image corresponding to a different image region from the second photographed image. The first is made of an image synthesizing means for fitting synthesized in the captured image or the reference image with Eject and. The photographing means is a digital camera.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
In FIG. 1 which shows the surveillance camera which is embodiment of this invention, the surveillance camera 10 consists of the digital camera 11 and the control apparatus 12 which controls this. The digital camera 11 is electrically connected to the control device 12 via a known cradle 13 and a USB cable 14. Reference numeral 15 denotes a LAN (local area network) code for connecting the control device 12 to an intranet (internal network). Note that the digital camera 11 includes, for example, a CCD having 3.4 million pixels, and can output image data of a captured image with the number of pixels of 2304 × 1536 pixels.
[0013]
The surveillance camera 10 is installed in a corner of a room where a safe is placed, for example. The photographing lens 11a of the digital camera 11 is directed to the door side so that a person entering the room can be photographed by opening the door of the room, and the focus and aperture are adjusted to the door and its peripheral part (see FIG. 3). ). Note that the room in which the surveillance camera 10 is installed is a room in which only a limited number of people can enter and leave the office.
[0014]
In FIG. 2 illustrating the electrical configuration of the monitoring camera 10, the control device 12 includes a controller 16, a ROM 17, a work memory 18, a sensor 19, a captured image buffer memory 20, a reference image memory 21, a background image memory 22, and an image comparison unit 23. , Foreground image memory 24, image composition unit 25, output image memory 26, and LAN card 27.
[0015]
The controller 16 controls the entire sequence of the monitoring camera 10 according to a program stored in the ROM 17. The work memory 18 is a memory used as a work area for the controller 16. The sensor 19 includes a light projecting unit 19a that emits light toward the door side and a light receiving unit 19b that receives reflected light. The sensor 19 detects a person entering from the door of the room and sends the detection signal to the controller. 16 The captured image buffer memory 20 is a memory that temporarily stores image data of a captured image captured by the digital camera 11.
[0016]
The reference image memory 21 stores image data of a reference image 28 (see FIG. 3) that is captured in a state where no one is in the room, that is, a detection signal is not input from the sensor 19 to the controller 16. Since the reference image 28 is not stored in the reference image memory 21 at the start of operation of the monitoring camera 10, the image data of the captured image first captured by the digital camera 11 is stored in the reference image memory 21 as image data of the reference image 28. Remembered.
[0017]
The background image memory 22 stores a captured image taken by the digital camera 11 using a detection signal from the sensor 19 as a trigger as a background image 32 (see FIG. 4). In the background image 32, the background of the door 29, the room 30, and the like is shown in focus with proper exposure, and the foreground that looks like a person in front is out of focus and underexposed.
[0018]
The image comparison unit 23 reads the image data of the background image 32 from the background image memory 22, and also reads the image data of the reference image 28 from the reference image memory 21, and detects a difference between the images in units of pixels. In the present embodiment, the difference in luminance value for each pixel is compared, and a different image area 33 that differs by a predetermined threshold value or more is detected from the background image 32.
[0019]
The controller 16 generates a mask image 34 covering the different image region 33 from the contour information of the different image region 33 detected by the image comparison unit 23 (see FIG. 5). This mask image 34 is generated in order to clarify the range of the different image region 33. The controller 16 controls the digital camera 11 to set a focus / photometry area 36 (see FIG. 6) at the center of the mask image 34, perform focusing and photometry, and perform photographing. The image data of the captured image thus obtained is stored in the foreground image memory 24 as image data of the foreground image 38.
[0020]
The foreground image 38 is sharply focused and has a proper exposure at substantially the same position as the different image area 33 (if the person moves from the captured image 32 until the foreground image 38 is captured). A person 39 is shown, and the background door 29 and the like indicated by dotted lines are blurred and overexposed.
[0021]
The image composition unit 25 cuts out the image of the person 39 from the foreground image 38 and inserts it into the mask image 34 of the background image 32 to create an output image 41 (see FIG. 7). The output image memory 26 stores image data of the output image 41.
[0022]
The LAN card 27 is connected to the intranet via the LAN code 15. On the other hand, a personal computer is installed in a management room which is a room different from the room 30, and this personal computer is connected to the intranet. By operating the personal computer, the output image 41 read from the output image memory 26 can be displayed on the liquid crystal monitor.
[0023]
The operation of the monitoring camera 10 configured as described above will be described. The controller 12 is always supplied with power, and the digital camera 11 is also supplied with power through the USB cable 14 and the cradle 13. The controller 16 controls the surveillance camera 10 according to a sequence program shown in the flowchart of FIG. 8 stored in the ROM 17.
[0024]
Immediately after powering on the monitoring camera 10, the image data of the reference image 28 is not stored in the reference image memory 21, so the first captured image is stored in the reference image memory 21 as the reference image 28.
[0025]
The digital camera 11 repeats photographing at intervals of 5 seconds, and the photographed image is temporarily stored in the photographed image buffer memory 20, then sent to the reference image memory 21, and overwritten and saved on the reference image 28. That is, unless a detection signal is input from the sensor 19 to the controller 16, the reference image 28 is always replaced with a new captured image at intervals of 5 seconds.
[0026]
When a detection signal is input from the sensor 19 to the controller 16, the controller 16 controls the digital camera 11 to release the shutter of the digital camera 11. The photographed image photographed by the digital camera 11 is temporarily stored in the photographed image buffer memory 20, then sent to the background image memory 22, and stored as the background image 32. Note that shooting at intervals of 5 seconds by the digital camera 11 is interrupted from when the detection signal is input to the controller 16.
[0027]
The controller 16 reads the image data of the background image 32 from the background image memory 22 and sends it to the image comparison unit 23, and also reads the image data of the reference image 28 from the reference image memory 21 and sends it to the image comparison unit 23. The image comparison unit 23 compares the background image 32 and the reference image 28 to detect a different image area 33.
[0028]
The controller 16 generates a mask image 34 from the contour information of the different image area 33 and covers the different image area 33 of the background image 32, and then controls the digital camera 11 to focus / photometric area at the center of the mask image 34. 36 is set to release the shutter of the digital camera 11.
[0029]
The digital camera 11 performs shooting after focusing and photometry using the focus / photometry area 36, and outputs image data of the captured image to the captured image buffer memory 20. The image data of the captured image is transferred from the captured image buffer memory 20 to the foreground image memory 24 and stored as image data of the foreground image 38.
[0030]
Thereafter, the image data of the background image 32 and the image data of the foreground image 38 are sent to the image composition unit 25, and the image of the person 39 corresponding to the mask image 34 is cut out from the foreground image 38 and the mask image 34 of the background image 32. It is inserted and synthesized. The image data of the composite image is stored in the output image memory 26 as image data of the output image 41.
[0031]
The image data of the output image 41 is sent from the LAN card 27 to the intranet, and the output image 41 is displayed on the liquid crystal monitor of the personal computer in the management room. Since both the person 39 and the room 30 shown in the output image 41 are clear, the administrator immediately determines whether or not the person 39 shown in the output image 41 is permitted to enter and leave the room 30. Can be determined. If the person is not permitted to enter or leave the room 30, appropriate measures such as prompt reporting to the police can be taken. In the case of an authorized person, a monitoring continuation command is sent from the personal computer in the management room to the control device 12 to resume photographing at intervals of 5 seconds.
[0032]
Since the output image 41 has, for example, 2304 × 1536 pixels, if it is output by a high-resolution ink jet printer or sublimation thermal transfer printer connected to a personal computer, it can be printed with an image quality very close to a photograph. This is effective for identifying the person 39 who has entered the room 30.
[0033]
The mask image is effective for setting the focus / photometry area and fitting / combining in order to clarify the range of the different image area. However, the foreground image is set by setting the focus / photometry area at the center of the different image area. The foreground image and the reference image are compared to detect a different area, the area is cut out from the foreground image, and this can be overwritten and combined with the reference image. In this case, generation of the mask image can be omitted.
[0034]
In the embodiment described above, the digital camera takes an image when the sensor detects a person who has entered the room, and stores the taken image as a background image. However, the interval is shorter than the above-described embodiment, for example, 1 Photographing may be performed at intervals of seconds, and the captured image and the reference image may be compared each time, and the captured image at the time when the different image area is detected may be stored as a background image. In this case, detection means such as a sensor can be omitted.
[0035]
Further, although the shooting interval in a state where there is no movement is 5 seconds, the present invention is not limited to this, and an arbitrary time such as 1 second or 10 seconds can be set. Further, instead of the sensor, a micro switch may be provided on the door, and the shutter of the digital camera may be released by a signal from the micro switch when the door is opened. Needless to say, the number of pixels of the captured image output from the digital camera is not limited to the above value.
[0036]
Further, the above embodiment is a monitoring camera that monitors a room in which a safe is installed, but the present invention is not limited to this, and for example, a monitoring camera is installed in a contracted elderly person's room. It may be a surveillance camera that monitors the elderly for health problems such as seizures.
[0037]
【The invention's effect】
As described above, according to the surveillance camera of the present invention, the first captured image captured by the imaging unit when an intruder is detected is compared with the reference image, and the focus position is set at the center of the detected different image area. And the photometry area are taken together, and the image corresponding to the different image area is cut out from the second photographed image and is inserted into the first photographed image or the reference image and synthesized, so that it corresponds to the difference image area of the second photographed image. Even when the shooting conditions of the image (foreground of a person or the like) and the first shooting image (background) are significantly different, both images can be automatically shot clearly. In addition, a difference image area which is changed by comparing a captured image captured by the imaging unit with a reference image is detected, and the focus position and the photometry area are aligned at the center of the difference image area. If the image corresponding to the different image region is cut out from the second photographed image and is inserted into the first photographed image or the reference image and synthesized, the detection means can be omitted and the object can be achieved. If the photographing means is a digital camera, the object can be achieved with a simple configuration.
[Brief description of the drawings]
FIG. 1 is a perspective view showing an external appearance of a surveillance camera.
FIG. 2 is a block diagram showing an electrical configuration of the surveillance camera.
FIG. 3 is an explanatory diagram showing a reference image.
FIG. 4 is an explanatory diagram showing a captured image.
FIG. 5 is an explanatory diagram showing a background image.
FIG. 6 is an explanatory diagram showing a foreground image.
FIG. 7 is an explanatory diagram showing an output image.
FIG. 8 is a flowchart showing an operation sequence of the surveillance camera.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 Surveillance camera 11 Digital camera 12 Control apparatus 16 Controller 21 Reference image memory 22 Background image memory 23 Image comparison part 24 Foreground image memory 25 Image composition part 26 Output image memory 28 Reference image 32 Background image 33 Difference image area 34 Mask image 36 Focus Photometry area 38 Foreground image 41 Output image

Claims (3)

An imaging unit that is installed at a predetermined position and that images a monitored location at regular intervals, a detection unit that detects an intruder into the monitored location, and an imaging unit when the detection unit does not detect an intruder A reference image storage means for storing image data of a photographed image taken by the camera as image data of a reference image, and image data of the photographed image photographed by the photographing means when the detecting means detects an intruder, as a first photographed image. A first photographed image storage means for storing the first photographed image data, an area detecting means for detecting a different image area that is different from the first photographed image compared with the reference image, and the photographing means to control the difference image. An imaging control unit that performs imaging by aligning the focus position and photometry area at the center of the area, and an image of the captured image that is captured by the imaging unit with the focus position and photometry area aligned at the center of the different image area Second photographed image storage means for storing data as image data of the second photographed image, and image composition means for cutting out an image corresponding to the different image area from the second photographed image and fitting and synthesizing the image into the first photographed image or the reference image. A surveillance camera characterized by comprising: An imaging unit that is installed at a predetermined position and captures a monitored place at regular intervals, a reference image storage unit that stores image data of a reference image previously captured by the imaging unit, and a constant by the imaging unit An area detection unit that detects a different image area that is changed by comparing a captured image that is captured every time with a reference image; and image data of the first captured image that is image data of the captured image in which the different image area is detected. A first photographic image storage means for storing the photographic image, a photographic control means for controlling the photographic means to shoot with a focus position and a photometric area in the center of the different image area, and a photographic control means for photographing. Second photographed image storage means for storing the image data of the photographed image as image data of the second photographed image, and extracting an image corresponding to the different image region from the second photographed image The first captured image or the monitoring camera, comprising the image synthesizing means for fitting synthesized reference image Te. The surveillance camera according to claim 1, wherein the photographing unit is a digital camera.

Images