JP2002157664A - Monitor system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a monitor system which provides service corresponding to a generated event. SOLUTION: This monitor system includes an observation part 3 which creates external field video by observing the external field, an image processing part 4 which extracts body video being video of an object body from the external field video, a judgment part 5 which judges an event related to the object body according to the body video, and a service providing part 2 which provides the service corresponding to the event.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a monitoring system, and more particularly to a monitoring system that understands an externally generated event and provides a service.

[0002]

2. Description of the Related Art Comprehensive events, such as the detection of a specified event occurrence, the evaluation of information on a specified event, and the prediction of the occurrence of a specified event, are monitored by humans by understanding the situation of the outside world. . Monitoring of external situations by humans
Over a long period of time, the observer becomes tired, and the level of monitoring and the judgment method vary due to the change of the observer. Furthermore,
In the case of a wide range, oversight of an event occurs due to concentration of attention. Further, for an event in which a large number of pieces of information are integrated and determined, an appropriate level of determination cannot be made unless a skilled person is used.

[0003] Monitoring systems that assist in these monitoring tasks are known. Such monitoring systems automatically discover only limited and very simple events. Furthermore, a monitoring system including an information processing device is particularly effective when a very high-speed sampling is performed to monitor an event that requires an instantaneous decision.

FIG. 17 shows a known monitoring system in a tunnel. In this monitoring system, as shown in FIG. 17, a fixed ITV camera 102 is installed and monitored at a considerable distance (normally every 200 m) in a tunnel 100. In addition, patrol vehicles patrol and monitor for safety. In such a monitoring system,
A falling object 101 near the middle point of the fixed ITV camera 102
When there is, as shown in FIG.
Is displayed on the screen in a small and unclear manner, and it is difficult to judge whether or not it will hinder the passage of the tunnel. Patrol vehicles patrol at considerable intervals in time, and it takes time for falling objects to be found and recovered.

[0005] In a welding operation using an inert gas, an observer monitors the welding operator and predicts an accident due to generation of harmful gas or lack of oxygen. Monitoring an event that occurs extremely infrequently deprives the observer of concentration and distracts the monitoring attention. Further, the worker is subjected to psychological resistance because the work status is monitored by the monitor.

A monitoring system in which various gas sensors are arranged in a work place to predict an accident due to generation of harmful gas or lack of oxygen is known. A gas sensor used in such a monitoring system detects only a specific gas, and does not detect all harmful gases that have an adverse effect on humans. There is known a monitoring method in which canaries are placed in a work place to predict the occurrence of oxygen deficiency or harmful gas. At this time, canaries respond to most gas accidents that have a negative effect on humans. In order to monitor canary, it is necessary to secure a dedicated observer, and the economic efficiency is low.

There has been proposed a next-generation city plan that centrally manages refuse in a downtown area in an underground storage and realizes efficient refuse collection while maintaining the beauty of the town. In such a city plan, it is necessary to make a plan to collect garbage when the amount of garbage increases, depending on the state of the garbage storage, and a monitoring system that constantly monitors the amount of garbage is desired.

[0008]

SUMMARY OF THE INVENTION An object of the present invention is to provide a monitoring system for providing a service corresponding to an event that has occurred. Another object of the present invention is to provide a monitoring system that efficiently provides a service corresponding to an occurred event. Still another object of the present invention is to provide a monitoring system that can efficiently find an event that has occurred. It is still another object of the present invention to provide a monitoring system that performs fine grain monitoring in time. It is still another object of the present invention to provide a monitoring system for monitoring a spatially fine grain of wood. Yet another object of the present invention is to provide a monitoring system for predicting danger. Still another object of the present invention is to provide a monitoring system that realizes comfortable living. Still another object of the present invention is to provide a monitoring system that realizes an elevator that operates efficiently. Still another object of the present invention is to provide a monitoring system for realizing a clean city. Still another object of the present invention is to provide a monitoring system that can be used for promoting an undeveloped area.

[0009]

Means for solving the problem are described as follows. The technical items appearing in the expression are appended with numbers, symbols, etc. in parentheses (). The numbers, symbols, and the like are technical items that constitute at least one embodiment or a plurality of the embodiments of the present invention, in particular, the embodiments or the examples. Corresponds to the reference numerals, reference symbols, and the like assigned to the technical matters expressed in the drawings corresponding to the above. Such reference numbers and reference symbols clarify the correspondence and bridging between the technical matters described in the claims and the technical matters of the embodiments or examples. Such correspondence / bridge does not mean that the technical matters described in the claims are interpreted as being limited to the technical matters of the embodiments or the examples.

A monitoring system according to the present invention comprises: an observation unit (3) for observing the outside world to generate an outside world image; an image processing unit (4) for extracting an object image which is an image of a target object from the outside world image; A determination unit (5) that determines an event related to the target object based on the video, and a service providing unit (2) that provides a service corresponding to the event. Observation unit (4)
Captures events directly or indirectly. The target object responds directly or indirectly to the event, and the reaction is captured as an external image by the observation unit (4).

The observing section (15) is fixed to a robot (13) running independently in a tunnel, the target object is a road in the tunnel, and the event is an obstacle when a vehicle runs in the tunnel ( 20, 21), and the service is preferably removal of obstacles (20, 21) or regulation of road traffic. The obstacle is a target object or an event, and includes, for example, a hole on a road.

Preferably, the event is the generation of a harmful gas or oxygen deficiency, the object is an organism (31) responsive to the generation of harmful gas or oxygen deprivation, and the service is a warning. Organism (31) is, for example, a canary that is more sensitive to humans and responds to the generation of harmful gases or oxygen deprivation.

[0013] The target object is waste (59), the event is that the waste (59) exceeds a predetermined amount, and the service is preferably the collection of waste (59). According to such a system, the waste (59) can be collected not at regular intervals, but as needed.

Preferably, the target object is a person waiting for an elevator, the event is a distribution of the number of persons on each floor, and the service is an elevator operation plan.

[0015] The observation unit is located in a nature conservation area, the target object is an animal, the event is that the animal is a rare species, and the service is the capture of an image of the rare animal. Is preferred. By arranging such a monitoring system in an undeveloped area, promotion of the area can be expected.

Preferably, the observation unit is arranged at the entrance of the management area, the target object is a person, the event is that the person is a suspicious person, and the service is that an alarm is issued.

The target object is colored so that the color difference from the background color is equal to or greater than a predetermined value.
It is preferable that the color is colored so that the color difference from the color of the target object is equal to or greater than a predetermined value, since the image processing unit (4) can easily extract the object image from the external image. Coloring is to change the color of the object to be colored, and is not limited to painting, but includes, for example, placing the target object in a bag of a predetermined color.

A pattern is arranged on the background of the target object to such an extent that the image processing unit can distinguish the object image from the background.
It is preferable that the target object has a pattern arranged on the surface of the target object to such an extent that the image processing unit can identify the background of the external image in that the image processing unit (4) can easily extract the object image from the external image.

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of a monitoring system according to the present invention will be described with reference to the drawings. The monitoring system includes an observation system 1 and a service providing unit 2, as shown in FIG. The observation system 1 includes an observation unit 3, an image processing unit 4, and a determination unit 5.

The observation unit 3 comprises a monitoring camera or a combination of a monitoring camera and a sensor other than the monitoring camera. The surveillance camera may be a fixed camera or a mobile camera, and may be single or multiple. The sensor can be fixed or movable, like the surveillance camera, and can be single or multiple. The surveillance camera shoots an image corresponding to a predetermined event, or
An image of an object responsive to the predetermined event is captured. The sensor observes air components, smells, weights, and the like that are not observed by the monitoring camera. Alternatively, a sensor is an object that is more sensitive to a given event than a human.

The image processing section 4 is mounted on an information processing apparatus exemplified by a personal computer. The image processing unit 4 identifies the target object and determines the position by performing image processing on the external image acquired from the monitoring camera of the observation unit 3.
The image processing unit 4 includes n (n = 1, 2, 3,...) Object extracting units 6-1 to 6-n, a background extracting unit 7, and n object identifying units 8-1 to 8-n. n and n object position determination units 9-1 to 9-1
9-n.

Each object extraction unit 6-i (i = 1, 2, 3,
, N) extracts an object image of the target object from the external image input from the monitoring camera of the observation unit 3. That is, the object extraction unit 6-i obtains the contour of the object by differentiation (difference) of the image, determines whether or not the object is a background pattern based on luminance near the contour. Is extracted as an object image. Alternatively, the object extracting unit 6-i extracts an object image by a moving stereoscopic view that three-dimensionally recognizes a target object and a background from a change in appearance when the monitoring camera moves. Or, the object extraction unit 6-i
Calculates an optical flow of an external image and extracts a portion of the motion pattern of the target object different from the background as a video of the object.

The background extracting unit 7 extracts a background portion from an external image in the same manner as the object extracting unit 6-i. At this time, it is preferable to color the color of the target object and the color of the background in a complementary color relationship in that the object image of the target object can be easily extracted. Specifically, the L of the color of the target object and the color of the background
It is preferable to increase the value such that ^* a ^* b ^* color difference ΔE ^* _ab can be determined by the observation unit 3 and the image processing unit 4. Only the target object or only the background may be colored. A special pattern may be arranged on the target object or the background so that the target object and the background can be easily distinguished. When it is difficult to extract the object by the image processing, the judgment may be made by a human.

Each object discriminating section 8-i has a database on the appearance of an object. Object identification unit 8
-I extracts an image near the contour of the object video extracted by the object extracting unit 6-i, and obtains and recognizes the similarity between the contour and the information recorded in the database. As such an identification method, an eigenspace method is known and is used for face recognition and the like. The object identification unit 8-i further generates a movement pattern (amount of movement per unit time) of the target object.

Each object position determining section 9-i determines the position of the target object based on the positional relationship between the image screen and the object image. In this position determination, which position on the acquired video screen corresponds to which actual position is registered in advance.
The position of the target object is determined based on where the object image is located on the image screen.

The determination unit 5 is mounted on an information processing device exemplified by a personal computer. The information processing device on which the determination unit 5 is mounted may be the same as or different from the information processing device on which the image processing unit 4 is mounted. In FIG. 1, the judgment unit 5 belongs to the observation system 1, but may belong to the service providing unit 2. The judgment unit 5 has a judgment table 11. Judgment table 1
Reference numeral 1 associates the occurred event acquired from the image processing unit 4 with the service content to be provided. The determination unit 5 passes the service content corresponding to the event that has occurred based on the determination table 11 to the service providing unit 2. Service provider 2
Provides a service to the user in response to an instruction from the determination unit 5.

FIG. 2 shows an example in which the monitoring system according to the present invention is used as a security confirmation system for confirming the safety in a tunnel. The safety confirmation system 12
As shown in FIG. 2, the inspection cart 13 and the monitoring center 1
4 The inspection trolley 13 corresponds to the observation unit 3. The inspection trolley 13 is a known robot that moves along a guide rail installed on the wall surface in the tunnel, and is used for cleaning the wall surface or inspecting equipment (lights and the like) in the tunnel.

The inspection trolley 13 has a monitoring camera 15,
A wireless transmitter 16 is provided. The surveillance camera 15 is fixed to the inspection trolley 13 so as to face the road surface and constantly photographs the road surface. The wireless transmitter 16 transmits an external image captured by the monitoring camera 15 to the monitoring center 14.
As the monitoring camera 15 moves, a falling object on the road is photographed closer, and a clearer image can be photographed. Further, the inspection cart 13 can monitor the inside of the tunnel more frequently than the patrol vehicle.

The monitoring center 5 includes a central control room 17, a display device 18 disposed in a tunnel, and a collection vehicle 19. The central control room 17 corresponds to the image processing unit 4 and the judgment unit 5 and includes an information processing device. The information processing device in the main control room 17 automatically analyzes an external image captured by the monitoring camera 15 and monitors whether an event causing a traffic accident has occurred. In the central control room 17, a monitor 22 belonging to the monitoring center 14 is arranged to prepare for an unexpected situation. The display device 18 and the collection vehicle 19 correspond to the service providing unit 2. The display device 18 is arranged in the tunnel or at the entrance of the tunnel, and notifies a passing vehicle of the presence of a falling object to warn.
The monitoring center 14 further dispatches a collection vehicle 19 to cause the collection vehicle 19 to collect the fallen objects.

FIG. 3 shows a danger judgment table 23 used in the safety confirmation system 12 and corresponding to the judgment table 11.
Is shown. The event that has occurred is retrieved from the judgment table 11, and the monitoring center 14 implements the contents of the service corresponding to the retrieved event. For example, when there is a cardboard box 20 on the road surface, the management center 14 immediately regulates the traffic on the lane where the cardboard box 20 is located and collects the collected vehicles 19.
To collect the cardboard box 20. If there is a puddle on the road surface, the management center 14 is left unattended.

The warning level may be changed according to the type and size of the falling object. For example, if there is an accident vehicle 21, traffic regulation is immediately performed, and if a falling object such as a string has little effect on traveling, it is not immediately recovered. When an event that is not described in the determination table 11 occurs, the monitor 22 may be asked to take specific measures by automatically saving the external image and alerting the monitor 22.

FIGS. 4 and 5 show the operation of the safety confirmation system 12. FIG. The external image captured by the monitoring camera 15 is transmitted from the inspection cart 13 to the monitoring center 14 by radio.
To the central control room 17. The information processing device in the central control room 17 performs known image processing on the acquired outside world video. First, a difference between the external image and the background luminance is extracted (step S1), and an image having a predetermined size or more is extracted from the extracted remaining image area (step S2). In the external image, the outline is extracted by differentiating the image (step S3). A portion where the extracted contours are concentrated is further extracted (step S4). In the external image, the moving pattern of the feature point in the moving image system is further measured (step S).
5). An image portion different from the road surface movement pattern is extracted (step S6). The common part of the three extracted videos is integrated into an object video. Alternatively, an object is identified using the individually extracted video as an object video (step S
7).

The outline is extracted from the object image (step S).
8) The extracted contour is compared with each object based on the contour feature database, and the similarity is calculated (step S9). The object video is further projected onto the eigenspace (step S10), and the similarity between each object and the feature point set in the eigenspace is calculated based on the database (step S11). The type of the target object is identified based on the calculated similarity (step S12).

The position of the object image in the visual field is extracted (step S13), and the actual position on the road corresponding to the position is calculated and determined (step S14). The information processing device in the central control room 17 collates the event information including the type and position of the object with the danger determination table 23 (step S15), and searches for service contents corresponding to the event information based on the danger determination table 23. I do. Upon acquiring the service content, the monitoring center 14 performs the service (step S17). For example, if a cardboard box is present in the vehicle lane, the monitoring center 14 displays the event information on the display device 18 and displays the collected vehicle 1
9. Collect the cardboard.

The monitoring of such a falling object prevents a vehicle traveling in the tunnel from suddenly braking and changing lanes in an attempt to avoid the falling object, thereby preventing an accident due to the sudden braking or the like. By moving the inspection trolley 13 and acquiring an image of a falling object or the like, as shown in FIG. 6, the object image 25 in the external image 24 becomes large and clear, and the determination of the type and position of the object becomes more difficult. Can be implemented in detail. Further, if the inspection cart 13 moves frequently at high speed, it is possible to realize fineness of spatial monitoring and fineness of temporal monitoring, and as a result, the reliability of the road is improved.

The inspection cart 13 is used only several times a year, and is not used for most of the period. Inspection trolley 13
Is more effectively utilized by being used in the monitoring system of the present invention. The inspection cart 13 is not limited to a robot in a tunnel, but may be an unmanned guided vehicle (AGV) disposed in a factory premises, a road safety inspection vehicle, a railway safety inspection vehicle before starting, or a construction machine capable of moving. It does not matter.

FIG. 7 shows an example in which the monitoring system according to the present invention is used in an accident occurrence prediction system for predicting an accident in a dangerous work area. Here, the dangerous work is a welding work using an inert gas or a work in a coal mine where there is a risk of landslides and falling rocks. The accident occurrence prediction system 30
Is a canary 31, a camera 32, a personal computer 33, a display device 34, and an alarm device 35
And

The canary 31 and the camera 32 are connected to the observation unit 3
It corresponds to. The canary 31 detects the generation of all harmful gases that have an adverse effect on humans, and detects oxygen deficiency. The camera 32 constantly shoots the canary 31 and transmits the image to the personal computer 33. The personal computer 33 corresponds to the image processing section 4 and the determination section 5 and observes the state of the canary 31 by performing image processing on a video acquired from the camera 32.

The personal computer 33 further has a danger judgment table 41 corresponding to the judgment table 11. FIG. 8 shows a specific example of the danger determination table 41. The danger determination table 41 associates the state of the canary 31 with the service content performed by the accident occurrence prediction system 30. The personal computer 33 displays the service content corresponding to the state of the canary 31 on the display device 34.
And the alarm device 35. In response to an instruction from the personal computer 33, the display device 34 blinks light or displays a message to notify the worker 36 that danger is predicted. The alarm 35 sounds a buzzer,
Alternatively, a message is sent to notify the worker 36 that the danger is predicted.

FIG. 9 shows an image generated by the camera 32. The image 38 is an object image 39 of the canary 31
And a background image 40. The background has a black and white striped pattern. In such a pattern, the background is easily distinguished from the canary 31 by image processing. The background may be colored with a complementary color of the color of the canary 31. Canary 3
1 may be colored to a complementary color of the background.

FIG. 10 shows the operation of the accident occurrence prediction system 30. The video taken by the camera 32
The data is transmitted to the personal computer 33. The personal computer 33 processes the acquired video by image processing.
The object video of the canary 31 is extracted, and the state of the canary 31 is identified. A background pattern is set in the personal computer 33 in advance, and the acquired video is compared with the background pattern, and different portions are extracted from the video (step S21). The personal computer 33 is
Further, the color of the canary is set in advance, and a portion similar to the color of the canary is extracted from the acquired video (Step S).
22). The personal computer 33 displays the extracted 2
Object image is generated by combining the two parts (step S
23).

The outline is extracted from the object image (step S2).
4), the aspect ratio of the contour is generated by shape matching or the like (step S25). The attitude of the canary 31 is generated from the aspect ratio of the contour. The moving amount of the object image with respect to the image acquired by the personal computer 33 is further generated (step S26). From this amount of movement, canary 3
One momentum is generated. The event information including the posture and the amount of exercise of the canary 31 is collated with the danger determination table 41 (step S27), and a service content corresponding to the event information is searched based on the danger determination table 41. The personal computer 33 makes the display device 34 and the alarm device 35 warn based on the searched service contents, and notifies the worker or the work supervisor that the danger is predicted (step S29).

FIG. 11 shows an example in which the monitoring system according to the present invention is used in a waste collection system for monitoring the amount of waste in a waste storage container. The refuse collection system 50 includes a container 51, a camera 52, an odor sensor 53, a weight sensor 54, and a refuse collection center 55, as shown in FIG.

The container 51 is buried underground in a downtown area and collects the refuse 59 discharged at the store. Container 51
Are arranged in a color or pattern that is easily distinguished from the dust 59. For example, the walls are colored black and the garbage bags are regulated to blue or white. The camera 52 is installed inside the container 51, photographs the inside of the container 51, and notifies the waste collection center 55. The odor sensor 53 detects a rotten odor generated from the refuse 59 and notifies the refuse collection center 55 of the odor. The weight sensor 54 measures the weight of the refuse 59 and notifies the refuse collection center 55 of the refuse.

The refuse collection center 55 includes a judgment table and a collection vehicle 57, and dispatches the collection vehicle 57 to the container 51 based on the information obtained from the camera 52, the odor sensor 53, and the weight sensor 54 and the judgment table. Garbage 5
9 is collected. The garbage collection center 55 further includes an observer 56, who prepares for an unexpected event.

The garbage collection system 50 is used to store garbage 5 in a downtown area.
9 are stored in a container 51 and centrally managed, and are collected when the amount of waste increases. FIG. 12 shows the operation of the refuse collection system 50. The image in the container 51 taken by the camera 52 is transmitted to the garbage collection center 55,
A difference from a preset background image is calculated (step S31), and if the remaining image area is equal to or larger than a predetermined size, the image is extracted (step S32). The video in the container 51 is further differentiated (Step S3
3) A portion where edges are concentrated is extracted (step S)
34).

The two extracted images are combined, and the amount of dust 59 is determined from the positional relationship between the known in-view image in the container 51 and the combined image (step S35). Garbage 5
If the amount of 9 exceeds the predetermined amount, the refuse collection center 55 dispatches the collection vehicle 57 to the container 51 and performs collection of the refuse 59. Further, if the odor sensor value obtained from the odor sensor 53 exceeds a predetermined value, or if the weight sensor value obtained from the weight sensor 54 exceeds a predetermined value, the dust 59 is similarly collected. (Step S38).

According to such refuse collection, a clean city and comfortable living are realized with high economic efficiency. It should be noted that if the object to be photographed is changed from the garbage in the container to the person in front of the elevator in such a monitoring system, it can be applied to the operation of the elevator depending on the number of people waiting.

FIG. 13 shows an example in which the monitoring system according to the present invention is used for an automatic detection system for rare species in a nature reserve. The automatic discovery system 60 includes a plurality of surveillance cameras 61-1 and 61-2 and a personal computer 6
2 and a recording device 63 for recording video. The plurality of monitoring cameras 61-1 and 61-2 correspond to the observation unit 3 and are arranged in a natural environment. The surveillance cameras 61-1 and 61-2 transmit the captured video to the personal computer 62 wirelessly.

The personal computer 62 corresponds to the image processing unit 4 and the judgment unit 5 and extracts living organisms from the scenery acquired from the surveillance camera 61, accurately identifies the extracted living organisms, and performs screening for recording them in a record. Judge your feelings. The recording device 63 corresponds to the service providing unit 2 and
This is a device for recording a video exemplified in a TR. The recording device 63 records the video transmitted from the personal computer 62.

The personal computer 62 includes a biological database 64 describing characteristics of the living thing and a judgment table 65 for determining whether the living thing is a rare or ordinary species.
And a video recording determination table 66. FIG.
An example of the video recording determination table 66 is shown.

FIG. 15 shows the operation of the automatic discovery system 60. The surveillance camera 61 captures and stores a background (landscape) video at predetermined time intervals (step S41). The captured background images are added to generate an average background image (step S42).

The image captured by the monitoring camera 61 is
The difference from the average background video is obtained (step S43). An area larger than a predetermined size is extracted from the remaining video (Step S44). The captured image is further subjected to an optical flow calculation to extract a motion pattern (step S45). An area that does not belong to a predetermined movement pattern is extracted by the optical flow (step S4).
6).

The common part between the object image extracted by the difference and the object image by the optical flow is integrated,
An object image is extracted (step S47). Alternatively, only the object video extracted by the difference or only the object video extracted by the optical flow may be extracted.

An outline is extracted from the extracted object image (step S48). The degree of similarity of the extracted contour is calculated from the feature of the contour based on the biological database 64 (step S49). The extracted object video is further projected on the eigenspace (step S50). The object image projected on the eigenspace is compared with a feature point set in the eigenspace of each living thing based on the living thing database 64, and the similarity is calculated (step S51). The two living things with the highest similarity are identified as the living things corresponding to the object (step S52).

The identified creature is compared with the video record determination table 66 (step S53), and it is determined whether to record or screen the video of the creature (step S54). If it is determined that the image should be recorded, the image of the living thing is recorded by the recording device 63 (step S5).
5).

According to such a surveillance system, images from a large number of surveillance cameras are automatically checked,
It is possible to effectively screen only the images of living things, and to help create a record of when and what kind of living thing was discovered. In addition, the use of this monitoring system in natural parks, native jungles, nature reserves, etc. facilitates the ecology of rare organisms. Alternatively, virtual insect collection or the like that captures valuable organisms as images may be performed. As a result,
Promotion of undeveloped areas can be expected.

Using such a monitoring system, an event with further enhanced entertainment may be held. For example, a participant carries a gun-type camera with which they can communicate, and goes out to nature-rich mountains and seas. If you find an unusual animal, shoot (capture) it with a gun-type camera. The captured video is sent to the counting center by wireless communication. Participants will be awarded points according to the rarity of the creature. Participants compete for how many points can be obtained within a predetermined time. By such an event, regional promotion can be expected. If the target is changed from a natural creature to a human, a suspicious person in a building or plant premises may be monitored.

[0059]

The monitoring system according to the present invention can provide a service corresponding to an event that has occurred.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of a monitoring system according to the present invention.

FIG. 2 is a block diagram showing a safety confirmation system.

FIG. 3 is a table showing a danger determination table;

FIG. 4 is a flowchart showing the operation of the safety confirmation system.

FIG. 5 is a flowchart showing the operation of the safety confirmation system.

FIG. 6 is a diagram showing an external image by the safety confirmation system.

FIG. 7 is a block diagram showing an accident occurrence prediction system.

FIG. 8 is a table showing a danger determination table.

FIG. 9 is a diagram showing an external world image by the accident occurrence prediction system.

FIG. 10 is a flowchart showing the operation of the accident occurrence prediction system.

FIG. 11 is a block diagram showing a refuse collection system.

FIG. 12 is a flowchart showing the operation of the refuse collection system.

FIG. 13 is a diagram illustrating a nature discovery system.

FIG. 14 is a table illustrating a video recording determination table;

FIG. 15 is a flowchart illustrating the operation of the nature discovery system.

FIG. 16 is a flowchart showing the operation of the nature discovery system.

FIG. 17 is a diagram illustrating an embodiment of a known monitoring system.

FIG. 18 is a diagram showing an external world image by a known monitoring system.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 ... Observation system 2 ... Service provision part 3 ... Observation part 4 ... Image processing part 5 ... Judgment part 6-1 to 6-n ... Object extraction part 7 ... Background extraction part 8-1-1 to 8-n ... Object identification part 9 -1 to 9-n: Object position determination unit 11: Judgment table 12: Safety confirmation system 13: Inspection trolley 14: Monitoring center 15: Camera 16: Wireless transmitter 17: Central control room 18: Display device 19: Collection vehicle 20 ... Falling object 21 ... Accident car 22 ... Observer 23 ... Danger judgment table 24 ... External image 25 ... Object image 30 ... Accident occurrence prediction system 31 ... Canary 32 ... Camera 33 ... Personal computer 34 ... Display device 35 ... Alarm 36 ... Worker 38 ... External image 39 ... Object image 40 ... Background 41 ... Danger judgment table 50 ... Garbage collection system 51 ... Container 52 ... Camera 53 ... Odor sensor 54 Weight sensor 55 ... Garbage collection center 56 ... Monitor 57 ... Collection vehicle 58 ... Store 59 ... Garbage 60 ... Nature discovery system 61-1 ... Surveillance camera 62 ... Personal computer 63 ... Recording device 64 ... Biological database 65 ... Judgment table 66 ... Video record judgment table 67 ... Researcher

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G08B 25/04 G08B 25/04 C H04N 7/18 H04N 7/18 DK F term (Reference) 3F303 CB22 CB31 5C054 AA01 EA01 EA05 EA07 FC01 FC03 FC12 FC13 FC14 FE28 GA00 HA00 HA01 HA18 HA26 5C084 AA02 AA07 AA13 BB33 CC17 DD11 EE01 EE03 EE04 FF02 FF03 DA27 FF43 GG43 GG52 GG57 GG78 HH03 AH13A12 CB27 FA02 FA06 5C087 AA02 AA03 AA19 AA22 AA32 AA37 AA42 AA44 BB03 BB11 BB18 BB74 DD03 DD14 DD18 DD23 DD25 DD26 EE05 EE16 FF01 FF02 FF04 FF17 FF19 FF20 FF22 GG02 GG06 GG20 GG20 GG20 GG20 GG20 GG20 GG20 GG20 GG20 GG20 GG20 GG20 GG20 GG23

Claims (11)

[Claims] An observing unit for observing the outside world to generate an external image, an image processing unit for extracting an object image as an image of the target object from the external image, and an image processing unit relating to the object based on the object image. A monitoring system comprising: a determining unit that determines an event to be performed; and a service providing unit that provides a service corresponding to the event. 2. The observation unit according to claim 1, wherein the observation unit is fixed to a robot that travels independently in a tunnel, the target object is a road in the tunnel, and the event is that a vehicle is traveling in the tunnel. A surveillance system, which is the presence of an obstacle when traveling, and the service is removal of the obstacle or regulation of traffic on the road. 3. The monitoring system according to claim 1, wherein the event is generation of harmful gas or oxygen deficiency, the target object is a living thing responding to the event, and the service is a warning. 4. The monitor according to claim 1, wherein the target object is waste, the event is that the waste exceeds a predetermined amount, and the service is collection of the waste. system. 5. The monitoring system according to claim 1, wherein the target object is a person waiting for an elevator, the event is a floor distribution of the number of persons, and the service is an operation plan of the elevator. . 6. The observation unit according to claim 1, wherein the observation unit is disposed in a nature protection area, the target object is an animal, the event is that the animal is a rare species, and the service is A surveillance system that captures an image of the rare animal. 7. The service according to claim 1, wherein the observation unit is arranged at an entrance of a management area, the target object is a person, the event is that the person is a suspicious person, and the service Is a monitoring system that is to issue an alarm. 8. The monitoring system according to claim 1, wherein the target object is colored so that a color difference from a background color becomes a predetermined value or more. 9. The monitoring system according to claim 1, wherein a background of the target object is colored so that a color difference from a color of the target object becomes a predetermined value or more. 10. The monitoring system according to claim 1, wherein a pattern is arranged on a background of the target object to such an extent that the image processing unit can identify the object image from the background. 11. The monitoring system according to claim 1, wherein the target object has a pattern arranged on a surface of the target object to such an extent that the image processing unit can identify the background of the external image.