JPH10108166A - Image monitor system with abnormality detection function - Google Patents

Abstract

PROBLEM TO BE SOLVED: To discriminate an abnormality from only an image picked up by a camera picking up a monitor area. SOLUTION: The system is provided with an image compression section 21 that compresses image data obtained from an image pickup section 1 such as a camera installed in a monitor area, a processing section 23 sets a threshold level to discriminate occurrence of a fault based on a data amount of the compressed image data obtained by the image compression section 21 and when there is a fluctuation in the amount of the compressed image data in excess of a prescribed range by using an abnormality didcrimination algorithm, it is discriminated that there is a change in the image of the monitor area picked up by the image pickup section 1 and an abnormality detection signal is outputted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image monitoring system for monitoring a monitoring area by an image captured by a camera for capturing a monitoring area for monitoring. The present invention relates to an abnormality detection device that detects a change in a captured image from a change in the amount of compressed image data obtained by compression, and a remote image monitoring system that uses the abnormality detection device.

[0002]

2. Description of the Related Art In a conventional image monitoring system for monitoring an image captured by an image capturing means, a monitoring operator determines whether an abnormality has occurred by looking at an image displayed on a monitor for monitoring, or captures an image of a monitoring area. The system is configured to detect an intrusion of a person by a sensor or the like provided in the vicinity of the camera and output an abnormality detection signal to notify a monitoring operator and transmit an image at that time to a monitoring image monitoring device. Was the most.

[0003]

SUMMARY OF THE INVENTION In a conventional image monitoring system, when an operator performs monitoring by watching and judging an image of a monitoring area displayed on a monitor,
The operator must constantly monitor the monitor, which makes it difficult to monitor the image for a long time, and may overlook or make a mistake in determination.

In addition, when a sensor for detecting an abnormality in a monitoring area is installed and an image is monitored in response to an abnormality detection signal from the sensor, a camera for photographing the monitoring area, a transmitting device for transmitting image data, and an image In addition to a monitoring device such as a monitor that displays the image, a sensor for detecting an abnormality in the monitoring area and a device for interlocking with the sensor are required, which complicates the system configuration.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problem, and an image is taken by a camera for taking an image of a monitoring area without using another sensor for detecting an abnormality such as an intruder in the monitoring area. An abnormality detection device that can judge anomalies only from images and an abnormality detection device that automatically determines an abnormality in the monitoring area and sends an abnormality detection signal and image data when an abnormality occurs to a remote image monitoring device It is an object of the present invention to provide an image monitoring system characterized by being able to do so.

[0006]

According to a first aspect of the present invention, there is provided an abnormality detecting apparatus for compressing image data obtained by compressing image data obtained from a camera installed in a monitoring area. Monitor changes in data volume,
When the data amount fluctuates by a predetermined amount or more, it is determined that a change has occurred in a captured image and an abnormality detection signal is output.

An abnormality detecting device according to a first aspect of the present invention is provided in an image monitoring system for providing an image capturing means for capturing an image of a monitoring area to be monitored, and for monitoring whether an abnormality has occurred in the monitoring area based on the captured image. In the abnormality detection device, image data input means for sequentially inputting image data of an image photographed by the photographing means, image compression means for performing data compression on the input image data, and image data compressed by the image compression means. From the data amount of the compressed image data obtained by setting the threshold value for determining the occurrence of an abnormality, and determining whether or not a change has occurred in the image due to a change in the data amount exceeding the threshold value. It is characterized by comprising processing means for outputting an abnormality detection signal.

In the image monitoring system according to a second aspect of the present invention, when an abnormality detection signal is output by the abnormality detection device according to the first aspect, the abnormality detection signal is transmitted to a remote image monitoring center. The image data is sent to the provided image monitoring device and the image data when an abnormality occurs.

According to a second aspect of the present invention, there is provided an image monitoring system, comprising: a photographing means for photographing a monitoring region to be monitored, and monitoring whether or not an abnormality has occurred in the monitoring region by the photographed image. Image data input means for sequentially inputting image data of an image photographed by the photographing means, image compression means for performing data compression on the inputted image data, and image data obtained by being compressed by the image compression means. A threshold for judging the occurrence of an abnormality is set based on the data amount of the compressed image data, and an abnormality detection signal is output which is characterized by a change in the image caused by a change in the data amount exceeding the threshold. An abnormality detection device comprising processing means; an image data storage means for storing compressed image data compressed by the image compression means; A terminal device comprising transmitting means for transmitting an abnormality detection signal output from the apparatus and compressed image data stored in the image data storage means to a remote image monitoring apparatus, and an abnormality detection signal and compressed image data transmitted from the terminal apparatus Receiving means for receiving compressed image data, compressed image data reducing means for reducing compressed image data transmitted from the terminal device to normal image data, and image data storing means for storing image data reduced by the compressed image data reducing means And display means for displaying image data transmitted from the terminal device or stored in image data storage means, and a notifying means for notifying a supervisor that an abnormality detection signal has been output from the terminal device, The image monitoring apparatus includes an input unit for a monitor to input an operation command and a processing unit for controlling each of the above units. It is characterized by a door.

[0010]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of an image sensing system according to the present invention; FIG. 1 is a block diagram illustrating a configuration of the abnormality detection device.

The imaging section 1 is composed of a CCD camera or the like for photographing the monitoring area, and outputs image data obtained by photographing the monitoring area to the abnormality judging section 2. The abnormality determination unit 2 includes an image input unit 21, an image compression unit 22, and a processing unit 23.

An image input unit 21 is an I / F for inputting image data taken by the image pickup unit 1 to the abnormality determination unit 2.
(Interface), which performs A / D conversion of the input image data and outputs the data to the image compression unit 22. The image compression unit 22 performs data compression to reduce the amount of image data input via the image input unit 21. The image compression unit 22 compresses image data using an image data compression algorithm described below. The compressed image data is created and output to the processing unit 23.

The processing unit 23 calculates the data amount of the compressed image data input via the image compression unit 22 and uses the abnormality determination algorithm described later to determine whether the data amount has exceeded a predetermined range. It determines that there is a change in the image of the monitoring area captured by the imaging unit 1, and outputs an abnormality detection signal. At this time, the compressed image data is output to the outside after the calculation of the data amount is completed.

With the above arrangement, the image data obtained by the imaging unit 1 is compressed by the image compression unit 22 and
It is determined from the change in the data amount of the compressed image data obtained by using the abnormality determination algorithm that a captured image has changed, and an abnormality detection signal is output.

When a large number of monitoring areas are monitored by sequentially switching cameras in a building or the like, the abnormality detection device of the present invention is applied, and an abnormality detection signal output from the abnormality determination unit 2 is used. Alternatively, the video may be switched to the video captured by the corresponding camera, and a warning sound may be emitted to notify the supervisor.

Further, the abnormality detecting device of the present invention is connected to an image or image data recording device such as a video tape recorder, a long-time recorded video, a magneto-optical disk, etc., and recording is started or output by outputting an abnormality detection signal. They may be linked so as to increase the frequency. Furthermore, in an image taken by a camera, a masking process of a known technology is performed for the purpose of reducing unnecessary noise or partially detecting an abnormality, so that unnecessary data is not included. Abnormality detection may be performed.

FIG. 2 shows the compression of the image data performed by the image compression section 22 of the abnormality determination section 2 as one of the image data compression methods.
JPEG (Joint Photograph)
FIG. 2 is a block diagram illustrating an image data compression algorithm when the image data compression is performed using the (icExperts Group) method.

The JPEG compression method is roughly divided into two stages, namely, frequency conversion by Fourier, quantization, and Huffman coding. Image data of the original image 221 captured by using a CCD camera or the like is input to the image input device 222 as each color component R, G, B, and thereafter, the luminance (Y)
And the color difference (C1, C2) components.

Next, for each of the Y, C1, and C2 components, 8 ×
The discrete cosine transform (hereinafter, DCT) is performed for each block with eight pixels as one block. The result obtained there,
Referring to a predefined quantization table 223, quantization (division) is performed so as to suppress high-frequency components, thereby reducing the data amount. Thereby, the first stage of the compression, that is, the operation of converting the spatial characteristics of the image data into the frequency characteristics and deleting the high-frequency component of the value to reduce the data amount is performed.

Next, Huffman coding is performed as a second stage of compression. In the actual Huffman coding, the direct current (DC) component and the alternating current (AC) component obtained in each block are coded. Here, only the DC component will be described. Briefly describing Huffman coding, it is a method of coding a difference between DC components of adjacent blocks. Natural images are CG (computer graphi)
Unlike cs), the average value between adjacent blocks does not change significantly, so that the difference between blocks concentrates around zero.

After performing the difference between the blocks, the compression is performed by giving a low bit data length to the statistically frequently used one by referring to the encoding table 224 for encoding. Data 225 is obtained. The image data is compressed by the above process. The data amount after compression becomes the same data amount if the captured image does not change, but the data amount after compression changes if the captured image is different. Therefore, for example, when the monitoring area is normal and the background image does not change, the data amount of the compressed image of the image captured by the imaging unit 1 is constant, but when a person or the like enters the monitoring area. Since the normal background image before the intrusion changes due to the intrusion of a person, the compressed image data of the captured image fluctuates. The present invention detects a change in the data amount of the compressed image data and detects an abnormality in the monitoring area.

The above description relates to the case of compressing image data of a color image, but the case of compressing image data obtained by photographing using a monochrome camera or an infrared camera. Similarly, if a change occurs in the captured image, the data amount of the compressed image data varies.

In the above description, JPEG is used as an image data compression algorithm. However, if an image data compression technique for performing compression processing such as DCT or Huffman coding is used, a compression due to a change in an image is performed. Since the amount of image data fluctuates to reach the output of the abnormality detection signal in the abnormality determination algorithm of the present invention, the compression technique is not limited to JPEG.

FIG. 3 is a flowchart of an abnormality determination algorithm used for abnormality detection in the processing section 23. The data amount of the compressed image data input via the image compression unit 22 in the processing unit 23 slightly fluctuates due to variations due to the imaging unit 1 such as a CCD camera. The variation due to the variation is defined as a threshold, and when the data amount of the compressed image exceeds the threshold, it is determined that the captured image has changed.

The abnormality determination algorithm is roughly divided into a pre-process (steps S001 to S005) and a detection process (steps S006 to S010). First, the pre-processing will be described. The pre-processing is for obtaining an initial value of the standard deviation σ necessary for performing the detection.

The reason for using the standard deviation σ is as follows.
Theoretically, if image data obtained by photographing a fixed image is input, the data amount of the compressed image is constant even if data compression is performed. However, in reality, noise generated in the imaging unit 1 due to environmental factors such as flickering of lighting in the monitoring area, and noise generated during A / D conversion in the image input unit 21 when inputting to the abnormality determination unit 2 For example, the input value of the compressed image data slightly fluctuates.

Further, since a large difference occurs in the data amount due to the background of the monitoring area of the object to be photographed, a fixed threshold cannot be determined. For the above reasons, the standard deviation σ is obtained in consideration of the fluctuation of the data amount of the compressed image data due to the noise input from the imaging unit 1 and the image input unit 21 described above, and the value is used as the threshold value.

The formula for obtaining the standard deviation σ is as follows.

[0029]

(Equation 1)

First, image data of an image photographed by the image pickup unit 1 is transmitted to an image compression unit 22 via an image input unit 21.
And data compression is performed (S002). The compressed image data compressed by the image compression unit 22 is input to the processing unit 23, and the processing unit 23 acquires the data amount Data (n) from the input compressed image data (S00).
3) The initial value of the standard deviation σ is calculated using the data amount Data (n) (S004). Here, it is determined whether or not the number N of data required for obtaining the initial value of the standard deviation σ has been obtained (S005). The process of S005 is repeated. On the other hand, when the specified number N is reached, the following detection processing is executed.

In the detection process, image data is input and compressed as in the preprocess (S006), and the data amount Data (n) of the compressed image data is obtained (S007). The data amount Da immediately before the obtained data amount Data (n)
The standard deviation σ is added to and subtracted from ta (n−1), and the two upper and lower thresholds and the data amount Data are obtained.
(N) is compared (S008). If the comparison result indicates that the data amount is within the range, the standard deviation σ is recalculated with the data amount Data (n) as the calculation target of the standard deviation σ, and the process returns to step S006 to repeat the processing of S006 to S008. If the result of the comparison is out of the range, it is determined that a change has occurred in the captured image, and an abnormality detection signal is output (S010).

In the above description, the standard deviation σ is used as it is as the threshold value. However, if it is desired to detect an abnormality with higher sensitivity, or to reduce false detection, the coefficient k is used as a sensitivity adjustment. The multiplied kσ may be used as the threshold. In this case, σ is replaced with kσ in step S008.

FIG. 4 shows the variation in the amount of compressed image data obtained by compressing the image data of a captured image.
Using an abnormality detection device that detects changes in the captured image,
FIG. 1 is a system configuration diagram when an image monitoring system is configured. In FIG. 4, a terminal device 4 is installed in each of a plurality of monitored objects 3. Each terminal device 4 detects an abnormality, and outputs an abnormality detection signal and compressed image data when the abnormality is detected.

The terminal device 4 includes a communication line 7 and an exchange 8
Is connected to an image monitoring device 6 provided in a remote image monitoring center 5 via an external device, so that an abnormality detection signal and compressed image data output from the terminal device 4 are sent to the image monitoring device 6. FIG. 5 shows a configuration of the terminal device 4 in the image monitoring system.

The terminal device 4 has a CCD for photographing a monitoring area.
An imaging unit 41 including a camera, an abnormality judging unit 42 that takes in image data, performs compression, detects abnormality, and the like, an image FIFO memory 43 and an image accumulation unit 44 that accumulates compressed image data, an abnormality detection signal, compressed image data, and the like. It comprises a transmission unit 45 for transmitting and a remote controller 46 for switching between starting and canceling monitoring.

In the basic operation, when an operation for starting monitoring is performed by the remote controller 46, image data of an image captured by the imaging unit 41 is transmitted to the image input unit 421 of the abnormality determination unit 42. The data is input to the image compression unit 422 and data compression is performed. The compressed image data compressed by the image compression unit 422 is sent to the processing unit 423 to determine the amount of data, and it is determined whether or not an abnormality has occurred based on the change in the amount of data using the abnormality determination algorithm described above.

The amount of data is obtained, and the compressed image data determined to be normal by the processing unit 423 is
The data is output to the IFO memory 43. On the other hand, the remote controller 46
The compressed image data when monitoring is started, the compressed image data at regular time intervals, and the compressed image data when the abnormality determination unit 42 determines that an abnormality has occurred are output to the image storage unit 44.

The transmission section 45 connects the terminal device 4 and the image monitoring device 6 provided in the image monitoring center 5 via the communication line 7 and the exchange 8, and transmits an abnormality detection signal. In addition, it reads out compressed image data or the like when an abnormality occurs from the image storage unit 44 or the image FIFO memory 43 and transmits it to the image monitoring device 6 provided in the image monitoring center 5. Further, it receives various signals such as a transmission stop signal output from the image monitoring device 6 of the image monitoring center 5 to the terminal device 4.

First, the monitoring operation in a normal state will be described. When the remote controller 46 is operated and monitoring is disclosed, the processing unit 423 sets the image input unit 421 and the image compression unit 4
The compressed image data at the start of monitoring, which is input via the interface 22, is stored in the image storage unit 44.

Thereafter, the image data is sequentially taken in, compressed by the image compression section 422, and whether or not an abnormality has occurred is determined by the processing section 423 using the above-described abnormality determination algorithm. , And outputs the compressed image data to the image FIFO memory 43. The image FIFO memory 43 discards the stored old compressed image data and updates it with new compressed image data until an abnormality detection signal from the processing unit 423 is received. When the terminal device 4 is connected to the image monitoring device 6 of the image monitoring center 5 via the communication line 7 and the exchange 8, the image FIFO memory 43 stores compressed image data until the terminal device 4 is connected. In addition, by storing compressed image data for a certain period of time, when an abnormality is detected, the image is checked immediately before the abnormality is detected and the situation is grasped. To do it.

The compressed image data being monitored is stored at regular intervals in the image storage section 44 as normal-time compressed image data. Are sequentially updated. The above operation is continued until the monitoring state is released by the remote controller 46.

Next, the operation when an abnormality is detected will be described. Abnormality judging section 42 by the above-mentioned abnormality judging algorithm
Has determined that an abnormality has occurred, the processing unit 423 temporarily stops accumulation of the compressed image data in the image FIFO memory 43 and stores the compressed image data at the time of occurrence of the abnormality in the image storage unit 44.

Thereafter, in order to transmit the compressed image data to the predetermined image monitoring device 6, an image transmission command is output to the transmission unit 45, and thereafter, the connection of the communication line 7 is confirmed. When the connection of the communication line 7 is confirmed, an abnormality detection signal for notifying that an abnormality has occurred is transmitted, and the data stored in the image FIFO memory 43 is transmitted in order from the oldest one.

When the compressed image data stored in the image FIFO memory 43 is transmitted to the image monitoring device 6 and the stored compressed image data decreases, the processing unit 423 sets the image F
The storage of the compressed image data in the IFO memory 43 is restarted,
The transmission of the compressed image data is continued until a transmission stop signal is received from the image monitoring device 6 or a predetermined time elapses.

At the time of connection of the communication line 7, if the remote image monitoring center 5 is in a busy state such as busy, the connection operation is performed again after a few seconds, and the image monitoring center 5 is connected.
Repeat the reconnection operation until the connection is made to the side. Further, when a call is made from the image monitoring center 5 during monitoring and the communication line 7 is connected, and an image transmission command is received,
The compressed image data stored in the image FIFO memory 43 is transmitted. When a still image transmission command is received, the compressed image data stored in the image storage unit 44 at the start of monitoring, during monitoring, and when an abnormality is detected is transmitted to the image monitoring device 6 of the remote image monitoring center 5. .

FIG. 6 shows the configuration of the image monitoring device 6 in the image monitoring system. The image monitoring device 6 includes a transmission unit 61, an image decompression unit 62, an image storage unit 63, a processing unit 64, a customer DB storage unit 65, a display unit 66, and an operation unit 6.
7, a speaker 68, and a peripheral device 69.

The transmission unit 61 receives the compressed image data and the abnormality detection signal transmitted from the terminal device 4 provided in the monitored object 3, and transmits the image data from the image monitoring device 6 to the terminal device 4.
To transmit various signals. The image decompression unit 62
The input compressed image data is restored from a compressed state to a normal state. The image storage unit 63 stores the input compressed image data. The processing unit 64 controls the entire image monitoring apparatus 6 and performs signal processing. Customer DB storage 6
Reference numeral 5 stores customer information to be searched when an abnormality detection signal is sent from the terminal device 4. The display unit 66 displays the input image data or the image storage unit 63.
Is displayed. Operation unit 67
, The operator inputs a command for operating the image monitoring device 6. The speaker 68 emits a warning sound to notify the operator when an abnormality detection signal is output by the terminal device 4 or the like. The peripheral device 69 includes a large-capacity removable disk such as a video printer, a video tape recorder, and a magneto-optical disk for outputting compressed image data stored in the image storage unit 63 or image data transmitted from a terminal device to the outside. Consists of

Next, the operation of the image monitoring device 6 will be described. Based on the ID information included in the abnormality detection signal sent when the terminal device 4 detects the abnormality,
With reference to the customer DB storage unit 65, the display unit 66 displays the customer information corresponding to the terminal device 4 that has output the abnormality detection signal, and outputs a warning sound from the speaker 68 to notify the operator of the occurrence of the abnormality. .

Thereafter, the compressed image data transmitted from the terminal device 4 is input to the image decompression unit 62 and the image storage unit 63 via the transmission unit 61, where the decompression and storage of the compressed image data are performed. The expanded image data is displayed on the display unit 66 by the processing of the processing unit 64.

A transmission stop signal for stopping the transmission of the compressed image data is output from the processing unit 64 when the operator of the image monitoring device 6 performs an end operation from the operation unit 67, and is transmitted through the transmission unit 61. It is sent to the terminal device 4. In addition, the moving image data stored in the image storage unit 63 can be viewed at an arbitrary time by the operator inputting a moving image reproduction command from the operation unit 67, and the image is displayed on the You can also take out as.

Further, the operator operates the terminal device 4 being monitored.
Can be requested to transmit image data. The procedure is as follows. By selecting the monitored terminal device 4 from the customer DB storage unit 65, the processing unit 64 causes the transmission unit 61 to connect the line to the target terminal device 4, and the terminal device 4
Transmits an image transmission command to the. The terminal device 4 which has received this command transmits the compressed image data to the image monitoring device 6, and after receiving the compressed image data accompanying the command, the image monitoring device 6 transmits the compressed image data in the same manner as described above. Decompression, accumulation, and display are performed.

In the above description, a warning sound is issued using the speaker 68 as a means for notifying the operator that an abnormality has been detected. However, a sound message including a voice warning or customer information may be provided. Alternatively, the abnormality may be notified by blinking a lamp or turning on an indicator lamp.

[0053]

Next, a specific embodiment of image compression using an abnormality detection device in an image monitoring system according to the present invention will be described. FIG. 7 shows the results of an experiment performed to show that when two different images obtained by the photographing unit are compressed, a difference occurs in the amount of data after compression depending on the image. 7A is an image of an outdoor scene, and FIG. 7B is an image of an indoor scene. The number of pixels of the obtained image is 160 ×
It is 120 pixels. Image data compression is JP
The EG method was used, and the compression ratio was set to 10 in this experiment. In the image (a), the data amount before compression was 23454 bytes, but the data amount after compression is 45454 bytes.
On the other hand, in the image (b), the data amount of the same 23454 bytes becomes 3520 bytes after compression, and a difference appears in the data amount after compression in the image (b).

FIG. 8 shows an experiment conducted to detect an abnormality using the abnormality detection device according to the present invention. In this experiment, an experiment was conducted in which a camera was installed at a T-junction in the corridor to detect a passing person. FIG. 8A shows the position of the camera, and FIG. A) A in the figure
The images of the camera when passing through points C and C and when passing through point B are shown.

FIG. 9 is a graph showing, from the image obtained from the camera shown in FIG. 8, the temporal change of the compressed image data and the threshold value obtained by the abnormality detection algorithm described in the flowchart of FIG. It is shown. The number of pixels of the image obtained by the camera is 160 × 120 pixels as in the above experiment, and the image compression is performed by the JPEG method. In the figure, a thick line indicates a change with time of the compressed image data amount, and a dotted line indicates a change with time of the upper and lower thresholds. The data amount indicates a relative value when the initial value is 1. In this case, the threshold value is set to 5σ, and the specified number N in calculating the initial value of the standard deviation σ is set to 10 for calculation.

In FIGS. 9A and 9B, the areas shown in FIGS. 9A and 9C show the case where the person is walking in the areas indicated by A and B in FIG. 8, ie, the camera shows only the background of the hallway where no person exists. The area (B) shows image data when the person is passing point B, that is, when the person is in the field of view of the camera.
The data amount also fluctuates in the areas (A) and (C) due to environmental factors due to fluctuations in brightness in the monitoring area and noise generated from the system itself. The standard deviation σ varies with the change in the amount of image data. However, since the calculation of σ is obtained by calculating based on data a predetermined number of times earlier, it appears after a predetermined delay time from the change in the compressed image data.

As shown in the graph of FIG. 9, the amount of compressed data when a person appears in the field of view of the camera fluctuates more rapidly than when there is no person, and this is detected by comparison with a threshold value. It becomes possible. From the abnormal experimental results,
An abnormality can be detected by monitoring the amount of compressed data of the image obtained from the photographing means. By applying this to an image monitoring system, an image monitoring system having an abnormality detection device can be easily provided. .

[0058]

According to the present invention, the image data obtained by photographing the monitoring area utilizes the fact that when the image changes, the data amount when the image data is compressed is changed. A threshold value is obtained based on the data amount at the time of data compression, and a change in a captured image can be determined based on whether or not a change in the data amount exceeding the threshold value has occurred.

For this reason, in image monitoring, it has conventionally been necessary to separately provide a sensor for detecting an abnormality in the monitoring area. On the other hand, monitoring is performed only by using image data obtained by photographing the monitoring area. Area abnormality detection becomes possible.
In addition, because it is configured to be able to detect anomalies only by using image data, even if an image monitoring system that does not have an existing anomaly detection function is to have an anomaly detection function,
It does not require construction work for installing sensors, etc., and can provide an abnormality detection function simply by inputting image data to the abnormality detection device of the present invention, which can quickly and easily improve the function of existing image monitoring systems. Can be measured.

Further, processing steps such as data compression for abnormality detection, setting of a threshold value, and comparison of whether or not a change in data amount exceeds a threshold value can be replaced with software processing. Rewrite the program to add or change functions, such as adding mask processing to image data for more reliable detection, changing the formula for calculating the standard deviation σ for sensitivity adjustment, and changing the data compression algorithm. Can be easily done.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an abnormality detection device.

FIG. 2 is a block diagram illustrating an algorithm of image data compression.

FIG. 3 is a flowchart of an abnormality determination algorithm.

FIG. 4 is a configuration diagram of an image monitoring system.

FIG. 5 is a block diagram illustrating a configuration of a terminal device of the image monitoring system.

FIG. 6 is a block diagram illustrating a configuration of a monitoring device of the image monitoring system.

FIG. 7 is a diagram illustrating a data amount after compression when two different images are compressed.

FIG. 8 is a diagram showing an arrangement diagram of cameras and an obtained image in an experiment of the abnormality detection device.

FIG. 9 is a diagram illustrating an example of compressed image data.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 imaging unit 2 abnormality determination unit 21 image input unit 22 image compression unit 221 original image 222 image input unit 223 quantization table 224 encoding table 225 compressed image data 23 processing unit 3 monitored object 4 terminal device 41 imaging unit 42 abnormality determination Unit 421 image input unit 422 image compression unit 423 processing unit 43 image FIFO memory 44 image storage unit 45 transmission unit 46 remote controller 5 image monitoring center 6 image monitoring device 61 transmission unit 62 image decompression unit 63 image storage unit 64 processing unit 65 Customer DB storage unit 66 Display unit 67 Operation unit 68 Speaker 69 Peripheral device 7 Communication line 8 Switch

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁶ Identification code FI G08B 25/00 510 H04N 5/225 C H04N 5/225 G06F 15/62 380

Claims (2)

[Claims] 1. An abnormality detecting device in an image monitoring system for providing an image capturing means for capturing an image of a monitoring area for monitoring, and for monitoring whether or not an abnormality has occurred in the monitoring area based on the captured image. Image data input means for sequentially inputting image data of a captured image; image compression means for performing data compression on the input image data; and data of compressed image data obtained by being compressed by the image compression means From the amount, from the processing means to set a threshold value for determining the occurrence of an abnormality, determine whether or not a change has occurred in the image due to a change in the data amount exceeding the threshold value, and output an abnormality detection signal Anomaly detection device. 2. An image monitoring system, comprising: a photographing means for photographing a monitoring area for monitoring; and monitoring whether or not an abnormality has occurred in the monitoring area by the photographed image. Image data input means for sequentially inputting image data, image compression means for performing data compression on the input image data, and abnormal data based on the amount of compressed image data obtained by compression by the image compression means. An abnormality detection device that sets a threshold value for judging occurrence, and a processing unit that outputs an abnormality detection signal characterized in that a change has occurred in an image due to a change in data amount exceeding the threshold value; Image data storage means for storing compressed image data compressed by the image compression means; an abnormality detection signal output from the abnormality detection device; A terminal device including a transmission unit that transmits the compressed image data stored in the image data storage unit to a remote image monitoring device; a reception unit that receives an abnormality detection signal and compressed image data transmitted from the terminal device; Compressed image data reduction means for reducing compressed image data transmitted from the device to normal image data, image data storage means for storing image data reduced by the compressed image data reduction means, transmission from the terminal device, Alternatively, a display unit for displaying the image data stored in the image data storage unit, a notification unit for notifying a monitor that an abnormality detection signal has been output from the terminal device, and the monitor inputting an operation command An image monitoring system comprising an input means for controlling the image forming apparatus and a processing means for controlling each of the above means. Beam.