JP2001292441A - Multi-point information supervisory system and method and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a multi-point information supervisory system that can reduce temporal and man-hour costs with respect to a supervisory job and a human error, decrease a time required for detection processing from image information and simply provide a wide range of supervisory area by the use of an existing network. SOLUTION: An operation computer, an information storage computer and a characteristic analysis computer are interconnected via a network to send/ receive information mutually. A camera system for the information storage computer is provided in the vicinity of the control computer and the characteristic analysis computer acquires image information of a user and information of a person identifier and stores them to a database. When the system receives a search request of a specific person, the characteristic analysis computer calculates a correlation value between the image information stored in the database and the image information of the camera system and informs the user about the image information with a high correlation. Furthermore, the system raises a warning on intrusion of a person other than the specific person.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is mainly used for an apparatus for monitoring a person. The present invention is used for 1) acquiring information of a person operating an input terminal or the like, 2) searching for a location of a designated person, 3) detecting the presence of a person other than the designated person. I do.

[0002]

2. Description of the Related Art Along with the recent high performance and low price of a CCD camera, a monitoring device using a camera device has rapidly spread. In addition, by expanding the information distribution infrastructure via the Internet and the like, it has become possible to acquire information on remote locations at low cost, and the need for monitoring techniques from remote locations using networks has become apparent.

At present, a monitoring system combining a plurality of camera devices and a computer device is sold, and most of them are provided with a large-capacity storage device, and accumulate images from the camera device as digital images.

Since the video information stored as a digital image is stored together with the information on the recording time and the recording location, it is possible to search for the required video information without replacing the video tape or the like.

[0005]

However, since only the time and recording location information is added to the recorded video information, the video is first selected from the time and the acquisition location related to the required video information and then visually checked. It is necessary to search for necessary places by confirmation.

Further, when the acquisition time and the photographing place cannot be specified as in the case of monitoring a suspicious individual, it is necessary to visually check all the images. However, in the case of specifying a person, there is also a method of performing a video search from similar image information such as clothing color. However, when the time and the shooting location are various, the time required for the processing increases.

In such a conventional monitoring system,
Since an operation of searching for a necessary place by visual confirmation is required, time and human costs increase. Also,
Since it is a visual check, it is difficult to avoid human errors such as oversight. Furthermore, when performing automatic detection from image information, the processing time also increases in proportion to the increase in time and shooting location.

SUMMARY OF THE INVENTION The present invention has been made in view of the above background, and has been developed to reduce the time and human cost and human error involved in monitoring work, and to reduce the time required for detection processing from image information. It is an object of the present invention to provide a point information monitoring device and method, and a recording medium. SUMMARY OF THE INVENTION An object of the present invention is to provide a multipoint information monitoring device and method and a recording medium that can easily provide a wide monitoring area using an existing network.

[0009]

A first aspect of the present invention is a multipoint information monitoring apparatus, which is characterized by detecting input means and detecting the end of an input operation to the input means. Operating device having a means for performing the operation, a display device, a camera device, a sensor for detecting the presence or absence of a person in the photographing range of the camera device, and the sensor detecting and not detecting the presence of the user. Means for respectively generating image information of both images photographed by the camera device, and means for temporarily storing these image information together with installation position information, photographing time information and detection information of the sensor of the camera device. An information storage computer; a unit for obtaining image information for each individual according to the image information stored in the temporary storage unit; a database for storing the image information; Means for calculating a correlation value between the image information stored in the means for temporarily storing the image information and image information whose shooting time information is different from the image information. The computer and the computer for feature analysis are connected via a network and include means for mutually transmitting and receiving information.

It is desirable that a plurality of the operation computers and the information storage computers are provided, and that these computers are connected via the network and have means for mutually transmitting and receiving information. Thus, a wide monitoring area can be easily provided by using an existing network.

As described above, in the present invention, computers for performing various processes are classified into respective categories of accumulation, analysis, and operation, and output information of the camera device and the sensor and operation input of the user are shared via a network, and the computer is sequentially operated. By adding the detection information together with the image information, the processing required for monitoring the multipoint information can be automatically performed, so that time and human cost and human error can be reduced. The main contents of multipoint information monitoring are, for example, 1)
Information of a person who operates the operation computer is acquired, 2) a location of the designated person is searched, and 3) presence of a person other than the designated person is detected.

The operation computer includes means for notifying the feature analysis computer of a detection result of the means for detecting the end of the input operation together with information on a person identifier of a user who has performed the input operation, and the camera device Is provided at a position that includes the user within the photographing range, the sensor is provided at a position that can detect the presence or absence of the user, and the unit that acquires image information for each individual is According to the detection result of the means for detecting the end, the sensor detects the presence of the user at a time preceding the input operation end detection time by a predetermined time from the input operation end detection time. A means for obtaining from the means for temporarily storing, a means for detecting difference information between the obtained two pieces of image information, and a means for storing the difference information in the database for each individual image information. It is desirable to provide a means for storing together with information of the person identifier of the user as.

That is, every time the end of the input operation to the operation computer is detected, the image information and the person identifier information of the user who has used the operation computer until immediately before the end are obtained, and this is obtained for each individual. Is stored in the database as image information indicating the characteristics of At this time, image information of only the user can be obtained by subtracting image information of only the background where the user has not been imaged from image information where the user has been imaged. The image information of the user alone is stored in the database as the image information indicating the characteristics of each individual together with the information of the person identifier of the user. The end detection of the input operation to the operation computer can be detected by, for example, not performing the input operation for a while.

The operating computer comprises means for receiving input of information on a person identifier via the input means, and means for notifying the information on the received person identifier to the computer for characteristic analysis. The computer, when the information of the person identifier is notified, a means for acquiring from the means for temporarily storing both image information captured by the camera device while the sensor is detecting and non-detecting the presence of a person. Means for detecting difference information between the acquired two pieces of image information, and means for calculating a correlation value between the difference information and image information corresponding to the information on the person identifier stored in the database. And a means for displaying image information of the camera device on the display means when a correlation value higher than a threshold value is calculated according to the calculation result. It may be configured to include and.

That is, when the user inputs the identifier of the person he / she wants to search into the operation computer, the information of the person identifier is notified to the feature analysis computer. The feature analysis computer that has received this notification first obtains the latest image information stored in the information storage computer. Furthermore, the image information of the person corresponding to the notified information of the person identifier stored in the own database is extracted.
Calculates the correlation value between the image information obtained from the information storage computer and the image information extracted from its own database, and when there is image information having a correlation value higher than the threshold value, stores the image information provided by the image information. The image information of the camera device of the operation computer is displayed on the operation computer of the user. Thereby, the user can obtain the latest image information of the person corresponding to the information of the person identifier input by the user.

[0016] The feature analysis computer comprises: means for obtaining from the means for temporarily storing both image information taken by the camera device while the sensor is detecting and not detecting the presence of a person; Means for detecting difference information between the two pieces of image information, and means for calculating a correlation value between the difference information and image information of a previously designated person stored in the database by means for calculating the correlation value, When a correlation value lower than the threshold value is calculated according to the calculation result, an alarm may be issued or the difference information may be displayed on the display means.

That is, when a person other than the person designated by the user is searched, a warning is issued or the person is specified. The user inputs an input from an operation computer or the like in advance. Then, a specific person is designated in the computer for feature analysis. The feature analysis computer extracts image information corresponding to the designated person from its own database, and calculates a correlation value between the image information and the image information acquired from the information storage computer. At this time, when a correlation value lower than the threshold value is calculated, an alarm is issued because image information of a person other than the designated person has been detected. This can be used to search for suspicious individuals.

Instead of issuing an alarm, image information of a person other than the designated person can be distributed to the display means and displayed. Further, the operation computer for distributing an alarm or image information can be arbitrarily set by the user.

As another mode of use, a user can directly give an instruction to the information storage computer from the operation computer to obtain arbitrary image information.

A second aspect of the present invention is an operation computer provided in the multipoint information monitoring device of the present invention. A third aspect of the present invention is an information storage computer provided in the multipoint information monitoring device of the present invention. A fourth aspect of the present invention is a feature analysis computer provided in the multipoint information monitoring device of the present invention.

According to a fifth aspect of the present invention, a computer device having predetermined hardware and predetermined basic software installed on the hardware is further installed on the computer device to operate the computer device according to the present invention. It is a machine-readable recording medium in which software is recorded as a device corresponding to a computer for use.

According to a sixth aspect of the present invention, a computer device having predetermined hardware and predetermined basic software installed on the hardware is further installed on the computer device so that the computer device according to the present invention can be used. This is a machine-readable recording medium in which software is recorded as a device corresponding to the storage computer.

According to a seventh aspect of the present invention, a computer device having predetermined hardware and predetermined basic software installed on the hardware is further installed on the computer device to thereby provide the computer device with the features of the present invention. This is a machine-readable recording medium on which software, which is a device corresponding to an analysis computer, is recorded.

An eighth aspect of the present invention is a multipoint information monitoring method, which is characterized by detecting the end of an input operation performed by a user on an operation computer, detecting the end of the input operation, and using the result. A camera device that notifies the feature analysis computer of the information of the person identifier of the user, and the feature analysis computer includes the user within the shooting range at a time that is a predetermined time before the input operation end detection time according to the detection result. Image information in which the user is captured and image information of a background in which the user is not captured, and detects difference information between the obtained two pieces of image information, and stores the difference information in a database. The information is stored together with information on the user's personal identifier as information indicating characteristics of each individual.

Alternatively, the user inputs the information of the person identifier to the operation computer, and the operation computer notifies the input information of the person identifier to the feature analysis computer. Upon receiving the notification, obtains image information of both image information of a person existing in the shooting range captured by the camera device and image information of no human being, and detects difference information between the obtained image information of the two. And
A correlation value between the difference information and the image information of the person corresponding to the information of the person identifier stored in the database is calculated. When a correlation value higher than a threshold is calculated according to the calculation result, the image of the camera device is calculated. The information is to be displayed on the display means of the operation computer.

Alternatively, the computer for characteristic analysis acquires from the camera device both the image information of the image information in which the person exists in the photographing range and the image information of the image in which the person does not exist, and calculates the difference between the acquired image information. Detects information, calculates a correlation value between the difference information and image information of a predetermined person stored in the database, and issues an alarm when a correlation value lower than a threshold is detected according to the calculation result. Or displaying the difference information.

As described above, according to the present invention, it is possible to reduce time and human costs and human errors involved in monitoring work, and to reduce the time required for detection processing from image information. Further, a wide monitoring area can be easily provided by using an existing network.

[0028]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The configuration of a multipoint information monitoring apparatus according to an embodiment of the present invention will be described with reference to FIGS. FIG. 1 is an overall configuration diagram of the multipoint information monitoring device. FIG. 2 is a block diagram of a main part of the operation computer. FIG. 3 is a block diagram of a main part of the information storage computer. FIG. 4 is a block diagram of a main part of the computer for feature analysis.

The present invention relates to a multipoint information monitoring apparatus, which is characterized by an input unit 1 and a screen saver for detecting the end of an input operation to the input unit 1 as shown in FIG. An operation computer 30 including an application unit 2 and a display unit 3, a camera device 4 as shown in FIG. 3, a sensor 5 for detecting the presence or absence of a person within a shooting range of the camera device 4, and An observation information acquisition unit 6 and a background image generation unit 7 that respectively generate image information of both images captured by the camera device 4 while the sensor 5 is detecting and not detecting the presence of the user; 4
An information storage computer 40 including an information storage unit 8 for temporarily storing the installation position information, the photographing time information, and the detection information of the sensor 5, and the image information stored in the information storage unit 8 as shown in FIG. A feature amount acquisition application unit 9 for acquiring image information for each individual according to the above, and a personal feature accumulation unit 1 which is a database for accumulating the image information
0 and a feature analysis computer 50 having a correlation value calculation unit 11 for calculating a correlation value between the image information and image information stored in the information storage unit 8 and having different shooting time information. The operation computer 30, the information storage computer 40, and the feature analysis computer 50 are connected to the network 6
The information transmission / reception unit 12, the stored information distribution unit 13, and the feature analysis transmission / reception unit 14, which are connected to each other via the H.0 and mutually transmit and receive information, are provided.

Further, as shown in FIG.
0 and a plurality of information storage computers 40 are provided. These computers are connected to each other via a network 60, and the information transmission / reception unit 12 and the stored information distribution unit 13 mutually transmit and receive information. The feature analysis computer 50 also transmits and receives information to and from each other by the plurality of operation computers 30 and information storage computers 40 and the feature analysis transmission / reception unit 14.

The multipoint information monitoring apparatus according to the embodiment of the present invention will be described in more detail by taking a designated person search system as an example. The sensor 5 and the camera device 4 are connected to the information storage computer 40, and the information storage computer 40, the feature analysis computer 50
The operation computer 30 is connected via a network 60. The operation computer 30 has installed therein software for operating as a network client. Network server software is installed in the information storage computer 40 and the feature analysis computer 50.

FIG. 2 is a schematic diagram of the operation computer 30.
The operation computer 30 includes a screen saver application unit 2, a person search client application unit 15,
Information transmitting / receiving unit 12, network interface unit 16
including. An input unit 1 and a display unit 3 are provided outside.
The screen saver application unit 2 starts when there is no input to the input unit 1 for a certain period of time, and transmits the user's last input time, user ID, and operation computer position information via the network interface unit 16 to the information transmitting / receiving unit. 12 is transmitted.

When the information of a person identifier (hereinafter, referred to as a person ID) is input to the input unit 1 by manual operation, the person search client application unit 15 transmits the information of the person ID to the network interface unit 1.
6 and transmitted by the information transmitting / receiving unit 12.

FIG. 3 is a schematic diagram of the information storage computer 40. The information storage computer 40 includes an observation information acquisition unit 6, an information storage unit 8, a background image generation unit 7, a storage information distribution unit 13, and a network interface unit 16. A camera device 4 and a sensor 5 are provided outside. Image information captured by the camera device 4 is stored in the information storage unit 8 via the observation information acquisition unit 6. At this time, the installation position information and the photographing time information of the sensor 5 and the camera device 4 and the detection information of the sensor 5 are added to the stored image information. The sensor 5 is an infrared sensor or an ultrasonic sensor, detects the presence or absence of a person in the photographing range of the camera device 4, and adds the detection result and detection time to the image information as detection information.

The background image generation unit 7 sequentially extracts images at a time when a person was not present on the screen from the detection information of the sensor and the image information of the camera device stored in the information storage unit 8, and the background at that time. The information is stored in the information storage unit 8 as an image. The image information stored in the information storage unit 8 is distributed via the network 60 by the stored information distribution unit 13 via the network interface unit 16.

FIG. 4 shows a schematic diagram of the computer 50 for characteristic analysis. The feature analysis computer 50 includes a personal feature storage unit 10, a database management unit 17, a person search server application unit 18, a feature amount acquisition application unit 9, a feature analysis transmission / reception unit 14, and a network interface unit 16. When receiving the information of the person ID from the screen saver application unit 2 of the operation computer 30, the feature amount obtaining application unit 9 obtains the characteristics of the designated individual from the information storage computer 40 and stores it in the personal characteristic storage unit 10. I do. The person search server application unit 18
When the information of the person ID is received from the person search client application unit 15 of the operation computer 30, the feature of the designated person ID is searched from the personal feature storage unit 10 via the database management unit 17.

Further, the network interface unit 1
6, the current image information is obtained from each information storage computer 40, and the image information that matches the previous feature is searched. The search result is returned to the operation computer 30 via the network interface unit 16.

Next, the operation of the multipoint information monitoring apparatus according to the embodiment of the present invention will be described with reference to FIGS. First, a sequence of processing for acquiring and storing personal characteristics by the screen saver application unit 2 and the characteristic amount acquisition application unit 9 will be described with reference to FIG. The operation computer 30 operates as a screen saver, calculates the time at which the user made the final input from the operation time and the screen saver activation set time, and features information on the time, the operation computer installation location, and the user ID of the user. It is distributed to the analysis computer 50. Here, it is assumed that the information on the operation computer installation location and the person ID is set in advance on the operation computer 30 [time, place, ID distribution]. Feature amount acquisition application section 9 of feature analysis computer 50
Then, the information of the time, the installation location of the operation computer, and the person ID are received, and the image information of the specified time is requested to the information storage computer 40 which is in charge of the operation computer [designated time image request]. The information storage computer 40 distributes the image information at the specified time to the feature analysis computer 50 [receiving the specified time image]. A request for a background image at a specified time is made to the information storage computer 40 near the operation computer installation location [designated time background image request]. The information storage computer 40 distributes the background image at the specified time to the feature analysis computer 50 [receives the specified time background image]. In the background difference image generation unit 19 of the feature analysis computer 50, based on the acquired image data, the image information “observed image” when the user is present and the image information “background image” at the time when the user is not present are obtained. Is calculated, and the existence area in the image of the user is specified. FIG. 6 shows the concept of this difference processing.

In the difference processing, the observed image is Aij, the value of the red component at the coordinates (i, j) of the observed image is arid, and the value of the green component at the coordinates (i, j) of the observed image is agi.
j, the value of the blue component at the coordinates (i, j) of the observed image is abij, the background image is Hij, and the coordinates of the background image are (i, j).
h) is the value of the red component in j), hgij is the value of the green component at coordinates (i, j) of the background image, hbij is the value of the blue component at coordinates (i, j) of the background image, and Rij is the person region image. The value of the red component at the coordinates (i, j) of the person region image is rrij, the value of the green component at the coordinates (i, j) of the person region image is rgij, and the value of the blue component at the coordinates (i, j) of the person region image is Value rbi
j, the X coordinate value of the image is i, the Y coordinate value is j, and the area determination threshold value is th.

[0040]

(Equation 1) As a result of this difference processing, if the difference from the background image is equal to or greater than th, the observed image is copied. Otherwise, an image to which −1 is substituted is generated.

The color histogram generation section 20 calculates a color histogram by performing the following operation on the person region Rij [color histogram detection]. Here, it is assumed that the storage variable of the color histogram is hn, the element number of the color histogram is n, the maximum possible value of the color component is cmax, and 0 is substituted into hn as an initial value.

[0042]

(Equation 2) According to this formula, n is uniquely determined by the color ratio of the red component, the blue component, and the green component in each pixel in the person area, and hn, which is a frequency value storage variable for that color, is incremented by one. This includes, for example, a red component, a blue component,
Assuming that each green component has 10 gradations (cmax = 10), n is a number in the range of 1 to 1000 because its maximum value is 10 × 10 × 10 = 1000, and the frequency value storage variable hn is h1 to h1. h1000 are prepared. Then, each pixel in the person area is assigned one number from 1 to 1000 according to the color ratio of the red component, the blue component, and the green component, and 1 is added to the frequency value storage variable hn corresponding to the number. Will be. Also, R _ij = −
When the value is 1, it is a pixel other than the person area, so that it is not targeted and the frequency value storage variable hn does not change. Therefore, if the whole of the person area is scanned and this calculation is performed, the frequency distribution (histogram) in the area for each color ratio in the person area is hn.
Can be obtained as each value of each array.

By this color histogram detection processing, the color frequency of the person area image is substituted for each of hn. hn is stored in the personal characteristic storage unit 10 in association with the user ID information as the characteristic data [save to the personal characteristic storage unit].

Next, the sequence of processing for searching for a specific target person from the information of the person ID in the person search client application unit 15 and the person search server application unit 9 will be described with reference to FIG. For the person search client application unit 15 of the operation computer 30,
The user inputs and distributes information on the person ID of the search target person via the input unit 1 (person ID designation). The person search server application unit 18 of the feature analysis computer 50 receives the information of the person ID and loads feature data that matches the information of the person ID from the personal feature storage unit 10 via the database management unit 17. The loaded histogram data is stored as h0n in the primary storage area of the feature analysis computer 50 [feature file selection and loading]. The image of the latest time is requested from all the information storage computers 40 from the feature analysis computer 50 [latest time image request]. All the information storage computers 40 deliver the latest time image information to the feature analysis computer 50 [receive the latest time image]. The feature analysis computer 50 requests all information storage computers 40 for the latest time background image [latest time background image request]. All the information storage computers 40 deliver the latest time background image to the feature analysis computer 50 [receive the latest time background image]. The background difference image generation unit 19 calculates the difference between the image information “observed image” at the time when the person was present and the image information “background image” at the time when the person was not present, from all the acquired image information. A background region in the image is specified [background difference processing]. [Equation 1] As a result of this difference processing, the observed image is copied when the difference from the background image is equal to or larger than th, and otherwise, an image substituted with −1 is generated.

The color histogram generation unit 20 calculates a color histogram of the person area Rij by the following operation [color histogram detection]. Here, the storage variable of the color histogram is hni, the element number of the color histogram is n, and the maximum value that the color component can take is cmax, the histogram of the i-th person area found from the entire image information. [Equation 2] The correlation value between hni and h0n is calculated, and the hni that takes the maximum correlation value is calculated [correlation value calculation]. If there is an hni equal to or larger than the predetermined threshold, the process proceeds to the [designated camera image request] process. If it does not exist, a search impossible message is transmitted to the person search client application unit 15 [correlation value> threshold]. The feature analysis computer 50 requests the image of the latest time from the information storage computer 40 having the maximum correlation value [designated camera image request]. The information storage computer 40 distributes the image information at the specified time to the feature analysis computer 50 [receives the specified camera image]. The designated camera image received from the feature analysis computer 50 is distributed to the person search client application unit 15 of the operation computer 30 and displayed on the display unit 3 [image display]. According to the above method, it is possible to automatically obtain desired image information and information on the location of a specific person only by specifying the information of the person ID.

Next, the sequence of processing for monitoring a suspicious individual in the person search client application unit 15 and the person search server application unit 18 will be described with reference to FIG. The person search server application unit 18 of the feature analysis computer 50 loads all user feature data from the personal feature storage unit 10. The loaded histogram data is stored in hid [feature file selection and loading]. Here, id is a person ID. The feature analysis computer 50 requests all the information storage computers 40 for the latest time image information [latest time image request]. All the information storage computers 40 deliver the latest time image information to the feature analysis computer 50 [receive the latest time image]. The feature analysis computer 50 requests all information storage computers 40 for the latest time background image [latest time background image request]. All the information storage computers 40 deliver the latest time background image to the feature analysis computer 50 [receive the latest time background image]. The background difference image generation unit 19 calculates a difference between the image “observed image” at the time when the person was present and the image “background image” at the time when the person was not present, from all the acquired image information, and [Background subtraction processing] for specifying an existing area in the interior. [Equation 1] If the difference from the background image is equal to or larger than th by this difference processing, the observed image is copied, and otherwise, an image substituted with −1 is generated.

The color histogram generation unit 20 calculates the color histogram of the person area Rij by the following operation [color histogram detection]. Here, the storage variable of the color histogram is hn1, the element number of the color histogram is n, and the maximum possible value of the color component is cmax. [Equation 2] Correlation values are calculated for all combinations of hni and hid, and hni less than or equal to a predetermined threshold value are calculated.
[Correlation value calculation]. If there is an hni equal to or smaller than a predetermined threshold, an alarm message is transmitted to the person search client application unit 15. If not, the process returns to the image request for the other information storage computer 40, and the same process is repeated [correlation value> threshold]. According to the above method, it is possible to detect an intrusion of a person other than the designated person and issue a warning to the user.

The user inputs the image information of the person such as an employee and the information of the person ID to be excluded from the target of the alarm in advance from the input unit 1 of the operation computer 30. Further, the operation computer 30 on which the alarm is displayed can be arbitrarily set by the user. In this case, the number of operation computers 30 may be one or plural.
Alternatively, the user can set so as to obtain image information of a suspicious individual instead of receiving an alarm.

As another example of use, the user can directly give an instruction to the information storage computer 40 from the operation computer 30 to obtain image information from an arbitrary camera device 4.

[0050]

As described above, according to the present invention,
It is possible to reduce the time and human costs and human error involved in the monitoring operation, and to reduce the time required for the detection process from the image information. Further, a wide monitoring area can be easily provided by using an existing network.

[Brief description of the drawings]

FIG. 1 is an overall configuration diagram of a multipoint information monitoring device according to an embodiment of the present invention.

FIG. 2 is a block diagram of a main part of an operation computer according to the embodiment of the present invention.

FIG. 3 is a block diagram of a main part of an information storage computer according to the embodiment of the present invention.

FIG. 4 is a block diagram of a main part of a feature analysis computer according to the embodiment of the present invention.

FIG. 5 is a flowchart showing a feature amount storing operation according to the embodiment of the present invention.

FIG. 6 is a conceptual diagram of a difference process.

FIG. 7 is a flowchart illustrating a person search operation according to the embodiment of the present invention.

FIG. 8 is a flowchart showing a suspicious person search operation according to the embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Input part 2 Screen saver application part 3 Display part 4 Camera device 5 Sensor 6 Observation information acquisition part 7 Background image generation part 8 Information storage part 9 Feature amount acquisition application part 10 Personal characteristic storage part 11 Correlation value calculation part 12 Information transmission / reception part 13 Stored information distribution unit 14 Feature analysis transmission / reception unit 15 Person search client application unit 16 Network interface unit 17 Database management unit 18 Person search server application unit 19 Background difference image generation unit 20 Color histogram generation unit 30 Operation computer 40 Information storage computer 50 Computer for feature analysis 60 Network

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 5C054 AA02 AA05 CA04 CC02 CE16 CH03 CH06 DA08 EA05 EA07 EG10 EH07 FB03 FC01 FC07 FC12 FC16 FE28 GB02 GB05 GB15 GB17 GC03 GD06 HA18 5L096 BA02 CA05 CA07 DA01 DA03 FA34 FA35 LA05

Claims (14)

[Claims] 1. An operation computer comprising: input means; means for detecting the end of an input operation on the input means; and display means; a camera device; and the presence or absence of a person within a shooting range of the camera device. And a means for respectively generating image information of both the camera device that is detecting and not detecting the presence of a user, and a device that detects the position of the camera device. An information storage computer comprising means for temporarily storing information and photographing time information and detection information of the sensor, and means for acquiring image information for each individual according to the image information stored in the means for temporarily storing; A database for storing the image information, and a correlation value between the image information and the image information stored in the temporary storage unit and having different shooting time information. A computer for feature analysis comprising means for calculating; and a means for connecting the operation computer, the information storage computer, and the feature analysis computer via a network and mutually transmitting and receiving information. A multipoint information monitoring device characterized by the following. 2. The multipoint system according to claim 1, wherein a plurality of the operation computers and the information storage computers are provided, and the computers are connected to each other via the network to mutually transmit and receive information. Information monitoring device. 3. The operation computer further comprises means for notifying the feature analysis computer of a detection result of the means for detecting the end of the input operation together with information on a person identifier of a user who has performed the input operation. The camera device is provided at a position that includes the user within the photographing range, the sensor is provided at a position that can detect the presence or absence of the user, and the unit that obtains image information for each individual is the input device. Both image information captured by the camera device when the sensor is detecting and not detecting the presence of a user at a time that is a predetermined time before the input operation end detection time according to the detection result of the means for detecting the end of the operation Means for temporarily acquiring the image information from the means for temporarily storing the image information, means for detecting the difference information between the acquired image information and the image information for the individual, and 2. A multipoint information monitoring apparatus according to claim 1, further comprising means for storing the information together with information on the user's person identifier as information. 4. The computer for operation includes: a unit for receiving input of information on a person identifier via the input unit; and a unit for notifying the computer for characteristic analysis of the information on the received person identifier. The analysis computer, when notified of the information of the person identifier, acquires from the means for temporarily storing both image information captured by the camera device while the sensor is detecting and not detecting the presence of a person. Means for detecting difference information between the acquired two pieces of image information,
Means for calculating a correlation value between the difference information and image information corresponding to the information of the person identifier stored in the database by means for calculating the correlation value, and a correlation value higher than a threshold value is calculated in accordance with the calculation result. 2. A multi-point information monitoring apparatus according to claim 1, further comprising: means for displaying image information of the camera device on the display means when the calculation is performed. 5. A means for acquiring from the means for temporarily storing both image information taken by the camera device while the sensor is detecting and not detecting the presence of a person, wherein the computer for feature analysis comprises: Means for detecting the difference information between the two pieces of image information, and means for calculating a correlation value between the difference information and the image information of a predetermined person stored in the database by the means for calculating the correlation value. 2. The multipoint information monitoring device according to claim 1, further comprising: a unit that issues a warning when the correlation value lower than the threshold value is calculated according to the calculation result, or displays the difference information on the display unit. 6. An operation computer provided in the multipoint information monitoring device according to claim 1. 7. An information storage computer provided in the multipoint information monitoring device according to claim 1. 8. A computer for characteristic analysis provided in the multipoint information monitoring device according to claim 1. 9. A computer corresponding to the operation computer according to claim 6, further installed on a computer having predetermined hardware and predetermined basic software installed on the hardware. A machine-readable recording medium on which software used as a device is recorded. 10. A computer device having predetermined hardware and predetermined basic software installed on the hardware, and further installed on the computer device to make the computer device correspond to the information storage computer according to claim 7. A machine-readable recording medium on which software to be used as a device for recording is recorded. 11. The computer according to claim 8, wherein said computer is further installed in a computer having predetermined hardware and predetermined basic software installed in said hardware. A machine-readable recording medium on which software to be used as a device for recording is recorded. 12. Detecting the end of the input operation of the user on the operation computer, and notifying the detection result and the information of the user's person identifier to the feature analysis computer, and the feature analysis computer Image information of the user photographed by the camera device including the user within the photographing range at a time that is earlier than the input operation end detection time by a predetermined time in accordance with the image information of the user and a background image in which the user is not photographed And acquiring difference information between the acquired two pieces of image information, and storing the difference information as information indicating characteristics of each individual in a database together with information on the person identifier of the user. Multipoint information monitoring method. 13. A user inputs information of a person identifier to an operation computer, and the operation computer notifies the input information of the person identifier to a feature analysis computer. Upon receiving the notification, obtains image information of both image information of a person existing in the shooting range captured by the camera device and image information of no human being, and detects difference information between the obtained image information of the two. Calculating a correlation value between the difference information and image information of a person corresponding to the information of the person identifier stored in the database; and when a correlation value higher than a threshold is calculated according to the calculation result, the camera device A multi-point information monitoring method, characterized by displaying the image information of (1) on a display means of the operation computer. 14. A feature analysis computer acquires, from a camera device, image information of both image information in which a person exists in the photographing range and image information in which no person exists, and calculates a difference between the acquired image information. Detects information, calculates a correlation value between the difference information and image information of a predetermined person stored in the database, and issues an alarm when a correlation value lower than a threshold is detected according to the calculation result. Or displaying the difference information.