Image processing apparatus, method for processing image, and image processing program

Abstract

To provide an image processing apparatus, a method for processing an image, and an image processing program which can detect an object efficiently and accurately with a relatively small number of cameras even if the objects are distributed.SOLUTION: An image processing apparatus 1 includes: a storage unit 24 configured to store a detection program determination table TB in which at least a shooting area ID which identifies a shooting area of a camera whose shooting area is changeable and a detection program ID which identifies a detection program PG for detecting a predetermined object from an image captured by the camera are associated with each other; and a detection unit 32 configured to perform processing of specifying the detection program PG corresponding to the shooting area of the camera by using the detection program determination table TB stored in the storage unit 24, and detecting the object from the image captured by the camera by using the specified detection program PG.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an image processing device, an image processing method, and an image processing program. [Background technology]

[0002] In recent years, camera-based surveillance systems have increasingly been used to automate security or inspection tasks. These surveillance systems simulate patrols by security guards or patrol inspections by inspectors by sequentially switching between multiple cameras or sequentially switching between the angle of view and direction of the same camera. Monitoring using such surveillance systems is called "camera patrol," "video patrol," or "preset patrol." Note that when switching between the angle of view and direction of the camera, a PTZ (pan / tilt / zoom) camera is used. Such surveillance systems sometimes use image processing devices that process images captured by the camera to detect objects contained in the images.

[0003] The following Patent Documents 1 and 2 disclose conventional image processing devices. Specifically, Patent Document 1 discloses an image processing device that can detect multiple objects, whether moving or stationary, from an input image and can respond to different requirements. Furthermore, Patent Document 2 discloses a system that can reduce the computational load of image recognition processing and can accurately obtain indicated values ​​even from analog instruments. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2010-226687 [Patent Document 2] Japanese Patent Application Laid-Open No. 2014-032039 Summary of the Invention [Problem to be solved by the invention]

[0005] However, many cameras are required to monitor or inspect all of the dispersed monitoring targets, inspection targets, and other objects. While it is possible to reduce the number of cameras by setting multiple objects to be placed within the field of view of one camera, the size of the object image relative to the entire image captured by the camera becomes smaller, making it difficult to accurately detect the object. Furthermore, when different types of objects are placed within the field of view of a camera, multiple types of image processing that are prepared in advance must be performed on the captured image, making it difficult to efficiently detect the objects.

[0006] The present invention has been made in consideration of the above circumstances, and aims to provide an image processing device, an image processing method, and an image processing program that can detect objects efficiently and accurately using a relatively small number of cameras, even if the objects are scattered. [Means for solving the problem]

[0007] In order to solve the above problem, an image processing device (1) according to a first aspect of the present invention includes at least a memory unit (24) that stores table information (TB, TB1) in which shooting area identification information that identifies the shooting area (A1, A2, B1 to B3) of a camera (A, B) whose shooting area is changeable and detection program identification information that identifies a detection program (PG) that detects a predetermined object from an image captured by the camera are associated with each other, and a detection unit (32) that uses the table information stored in the memory unit to identify the detection program corresponding to the shooting area of ​​the camera and performs processing to detect the object from an image captured by the camera using the identified detection program.

[0008] Furthermore, an image processing device according to a second aspect of the present invention is an image processing device according to the first aspect of the present invention, further comprising a control unit (31) that instructs the camera to set the shooting area of ​​the camera to a predetermined shooting area, and the detection unit identifies the detection program using the table information based on the shooting area identification information that identifies the predetermined shooting area set by the control unit.

[0009] Furthermore, an image processing device according to a third aspect of the present invention is an image processing device according to the second aspect of the present invention, wherein the memory unit stores at least shooting time information that specifies the time at which shooting is performed by the camera and shooting schedule information (SC) that corresponds to the shooting area identification information, and the control unit issues the instruction based on the shooting schedule information stored in the memory unit.

[0010] In addition, an image processing device according to a fourth aspect of the present invention is an image processing device according to the first aspect of the present invention, wherein the detection unit, when the camera sets the shooting area to a predetermined shooting area, identifies the detection program using the table information based on the shooting area identification information that identifies the predetermined shooting area transmitted from the camera.

[0011] Furthermore, an image processing device according to a fifth aspect of the present invention is an image processing device according to the fourth aspect of the present invention, which includes a control unit (31) that causes the camera to set shooting schedule information in which shooting time information specifying the time at which shooting is performed by the camera is associated with the shooting area identification information.

[0012] Furthermore, an image processing device according to a sixth aspect of the present invention is an image processing device according to the third or fifth aspect of the present invention, wherein the table information is information in which camera identification information that identifies the multiple cameras, the shooting area identification information, and the detection program identification information are associated with each other, the shooting schedule information is information in which the shooting time information, the camera identification information, and the shooting area identification information are associated with each other, and the detection unit and the control unit identify the camera using the camera identification information.

[0013] Furthermore, an image processing device according to a seventh aspect of the present invention is an image processing device according to any one of the first to sixth aspects of the present invention, in which the detection unit performs preprocessing on the image taken by the camera before performing processing to detect the object from the image taken by the camera using the identified detection program.

[0014] Furthermore, an image processing device according to an eighth aspect of the present invention is an image processing device according to the seventh aspect of the present invention, wherein the table information is information in which the shooting area identification information, the detection program identification information, and preprocessing identification information that identifies the preprocessing are associated with each other, and the detection unit uses the table information to perform the preprocessing according to the shooting area of ​​the camera.

[0015] Furthermore, an image processing device according to a 9th aspect of the present invention is an image processing device according to the 3rd or 5th aspect of the present invention, wherein the control unit controls the camera to change the shooting conditions according to the image captured by the camera.

[0016] Furthermore, an image processing device according to a 10th aspect of the present invention is an image processing device according to the 3rd or 5th aspect of the present invention, in which the control unit changes the shooting time information depending on the state of the object detected by the detection unit.

[0017] An image processing method according to one aspect of the present invention includes a first step in which a memory unit (24) stores table information (TB, TB1) in which at least shooting area identification information identifying the shooting area (A1, A2, B1 to B3) of a camera (A, B) whose shooting area is changeable is associated with detection program identification information identifying a detection program (PG) that detects a specified object from an image captured by the camera, and a second step (S14) in which a detection unit (32) uses the table information stored in the memory unit to identify the detection program corresponding to the shooting area of ​​the camera and performs processing to detect the object from the image captured by the camera using the identified detection program.

[0018] An image processing program according to one embodiment of the present invention causes a computer to execute a first step of storing table information (TB, TB1) in which at least shooting area identification information identifying the shooting area (A1, A2, B1 to B3) of a camera whose shooting area is changeable and detection program identification information identifying a detection program (PG) that detects a specified object from an image captured by the camera are associated with each other, and a second step (S14) of using the table information to identify the detection program corresponding to the shooting area of ​​the camera and performing processing to detect the object from an image captured by the camera using the identified detection program. [Effects of the Invention]

[0019] According to the present invention, it is possible to detect objects efficiently and accurately using a relatively small number of cameras even if the objects are dispersed. [Brief explanation of the drawings]

[0020] [Figure 1] 1 is a diagram showing the configuration of a monitoring system in which an image processing device according to an embodiment of the present invention is used; [Figure 2] 1 is a block diagram showing the configuration of a main part of an image processing apparatus according to an embodiment of the present invention; [Figure 3]FIG. 10 is a diagram showing an example of a detection program determination table according to an embodiment of the present invention. [Figure 4] FIG. 10 is a diagram showing an example of shooting schedule information according to an embodiment of the present invention. [Figure 5] 4 is a flowchart illustrating an example of the operation of the image processing device according to the embodiment of the present invention. [Figure 6] 10 is a flowchart illustrating another example of the operation of the image processing device according to the embodiment of the present invention. [Figure 7] FIG. 10 is a diagram showing an example of a detection program determination table according to a modified example of an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0021] An image processing device, an image processing method, and an image processing program according to an embodiment of the present invention will be described in detail below with reference to the drawings. First, an overview of the embodiment of the present invention will be described, followed by a detailed description of the embodiment of the present invention.

[0022] 〔overview〕 The present invention provides an image processing device, an image processing method, and an image processing program that can efficiently and accurately detect objects using a relatively small number of cameras, even if the objects are scattered. The present invention can be applied to, for example, reading multiple types of meters in a plant or factory, or monitoring the water level of a river and the riverbed (for example, detecting abandoned garbage).

[0023] In a camera-based surveillance system, multiple different types of monitoring, inspection, or other objects may be placed within the field of view of a single camera. For example, in plants, factories, and other facilities, multiple different types of meters may be placed within the field of view of a single camera in order to use a camera to read the instrument readings.

[0024] There are various types of gauges, such as round, rectangular, level, and 7-segment LED (Light Emitting Diode) types, and the logic for detecting the gauge's reading from the image of the gauge (for example, the AI ​​(Artificial Intelligence) processing content) also differs for each type of gauge. When multiple different types of gauges are placed within the field of view of a single camera, the gauge's reading must be detected using multiple types of logic from the image captured by the camera, making it impossible to detect the target object efficiently.

[0025] An embodiment of the present invention stores table information in which at least photography area identification information and detection program identification information are associated with each other. The photography area identification information is information that identifies the photography area of ​​a camera whose photography area is changeable. The detection program identification information is information that identifies a detection program that detects a predetermined object from an image captured by the camera. An embodiment of the present invention uses the table information to identify a detection program corresponding to the photography area of ​​the camera, and performs processing to detect objects from images captured by the camera using the identified detection program. As a result, objects are detected for each photography area of ​​the camera using an appropriate detection program corresponding to the photography area of ​​the camera, so that even if objects are dispersed, it is possible to detect objects efficiently and accurately with a relatively small number of cameras.

[0026] [Embodiment] <Surveillance System> Fig. 1 is a diagram showing the configuration of a monitoring system in which an image processing device according to an embodiment of the present invention is used. As shown in Fig. 1, the monitoring system SY includes cameras A and B and an image processing device 1 that are communicatively connected via a network N. The network N is, for example, a LAN (Local Area Network) installed in a plant, factory, or other facility. The network N may be any of a network capable of wired communication, a network capable of wireless communication, and a network capable of both wired and wireless communication.

[0027] Cameras A and B are cameras whose shooting areas can be changed. Cameras A and B are, for example, PTZ cameras that can pan, tilt, and zoom remotely. Panning refers to swinging in the horizontal (left and right) direction, tilting refers to swinging in the vertical (up and down) direction, and zooming refers to enlarging or reducing the captured image. Cameras A and B can set (preset) one or more shooting areas. The shooting areas are set by, for example, a user who uses the image processing device 1. Note that cameras A and B may be capable of only one or two of panning, tilting, and zooming. Furthermore, the number of cameras is not particularly limited and may be one, three, or more.

[0028] For ease of understanding, it is assumed that multiple instruments (objects) are placed within the area that can be photographed by cameras A and B, and multiple photographing areas are set within the area that can be photographed by cameras A and B. In the example shown in Fig. 1, photographing areas A1 and A2 are set for camera A, and photographing areas B1 to B3 are set for camera B.

[0029] In the example shown in FIG. 1, the same type of instrument (for example, a circular one-needle meter) is placed in the photographing area A1 of camera A and the photographing area B1 of camera B. Furthermore, another same type of instrument (for example, a circular two-needle meter) is placed in the photographing area A2 of camera A and the photographing area B3 of camera B. Another type of instrument (for example, a level meter) is placed in the photographing area B2 of camera B. Note that the position, number, and size of the photographing areas are arbitrary, and the type of instrument placed in each of the photographing areas is also arbitrary.

[0030] The image processing device 1 has, as its functions, a communication control unit 11, a signal processing unit 12, and an output unit 13, and acquires images taken by cameras A and B via a network N, performs necessary processing on the acquired images, and then outputs the processing results. The communication control unit 11 acquires images taken by cameras A and B via the network N, and controls cameras A and B via the network N. Here, examples of controlling cameras A and B include setting the shooting area, setting the shooting schedule, controlling the shooting timing, and changing the shooting conditions.

[0031] The signal processing unit 12 performs necessary processing on the images acquired by the cameras A and B via the network N. For example, the signal processing unit 12 performs processing to detect instruments from the images captured by the cameras A and B, and performs processing to obtain the indicated values ​​of the detected instruments. The output unit 13 outputs the images acquired by the communication control unit 11, the processing results of the signal processing unit 12, etc. The output unit 13 displays, for example, the images acquired by the communication control unit 11, the processing results of the signal processing unit 12, etc., and outputs these data to the outside.

[0032] Image Processing Device Fig. 2 is a block diagram showing the main components of an image processing device according to one embodiment of the present invention. As shown in Fig. 2, the image processing device 1 includes an input unit 21, a display unit 22, a communication unit 23, a storage unit 24, and a processing unit 25. Such an image processing device 1 may be provided as a standalone device, or may be incorporated into a communication device such as a gateway or a router. Furthermore, the image processing device 1 may be realized by a computer such as a personal computer or a workstation.

[0033] The input unit 21 outputs instructions (instructions for the image processing device 1) to the processing unit 25 in response to operations by a user using the image processing device 1. The input unit 21 may be, for example, a switch such as a DIP switch, or an input device such as a keyboard or pointing device connected to the image processing device 1 via a wired or wireless connection. The display unit 22 displays various information output from the processing unit 25. The display unit 22 may include, for example, a display device such as an LED or a liquid crystal display device. The display device may be connected to the image processing device 1 via a wired or wireless connection. The input unit 21 and the display unit 22 may be physically separated, or may be physically integrated, such as a touch panel liquid crystal display device that combines display and operation functions.

[0034] The communication unit 23 communicates with the cameras A and B or terminal devices TM such as personal computers connected to the network N (see FIG. 1 ) under the control of the processing unit 25. The communication unit 23 transmits various information (e.g., control information or reports described later) output from the processing unit 25 to the cameras A and B or the terminal devices TM via the network N. The communication unit 23 receives various information (e.g., captured images, identification information of the cameras and the image capturing areas, or various data created by the terminal devices TM) transmitted from the cameras A and B or the terminal devices TM via the network, and outputs the information to the processing unit 25.

[0035] The storage unit 24 includes an auxiliary storage device such as a volatile or non-volatile semiconductor storage device, HDD (hard disk drive), or SSD (solid state drive), and stores various information. Specifically, the storage unit 24 stores the imaging data DT, the detection program PG, the detection program determination table TB (table information), and the imaging schedule information SC. The storage unit 24 may also store various programs that realize the functions of the image processing device 1, data that is temporarily used for the processing of the processing unit 25, and the like.

[0036] The photographing data DT is data of images photographed by the cameras A and B, and is data acquired by the image processing device 1 from the cameras A and B via the network N. Taking into consideration the capacity of the storage unit 24, the photographing data DT may be images photographed by the cameras A and B from the time when the latest image was photographed up to a predetermined time period prior to that time.

[0037] The detection program PG is a program that detects a predetermined instrument (predetermined object) from the image captured by cameras A and B and obtains the indicated value of the detected instrument. This detection program PG is prepared for at least each type of instrument captured by cameras A and B. For example, in the example shown in Figure 1, a detection program PG for a "circular one-needle meter," a detection program PG for a "circular two-needle meter," and a detection program PG for a "level meter" are prepared.

[0038] The detection program determination table TB is a table in which camera IDs (camera identification information), photographing area IDs (photographing area identification information), and detection program IDs (detection program identification information) are associated with each other. The camera IDs are information for identifying cameras A and B, the photographing area IDs are information for identifying photographing areas A1, A2, B1 to B3, and the detection program IDs are information for identifying detection programs PG. The detection program determination table TB is created, for example, by a user of the image processing device 1 operating the terminal device TM or the input unit 21. The created detection program determination table TB is stored in the storage unit 24 (first step).

[0039] 3 is a diagram showing an example of a detection program determination table in one embodiment of the present invention. In the example shown in FIG. 3, the camera ID of camera A is indicated as "AAAAA" and the camera ID of camera B is indicated as "BBBBB." Note that, for example, a MAC (Media Access Control) address or the like can be used as the camera ID. Also, the shooting area IDs of the shooting areas A1 and A2 shown in FIG. 1 are indicated as "A-1" and "A-2," respectively, and the shooting area IDs of the shooting areas B1, B2, and B3 are indicated as "B-1," "B-2," and "B-3," respectively. Note that any identification information can be used as the shooting area ID.

[0040] 3, the detection program ID of the detection program for the "circular one-needle meter" is represented by "α", the detection program ID of the detection program for the "circular two-needle meter" is represented by "β", and the detection program ID of the detection program for the "level meter" is represented by "γ".

[0041] 3, "AAAAA", "A-1", and "α" are associated with each other, and "AAAAA", "A-2", and "β" are associated with each other. Also, "BBBBB", "B-1", and "α" are associated with each other, "BBBBB", "B-2", and "γ" are associated with each other, and "BBBBB", "B-3", and "β" are associated with each other.

[0042] The shooting schedule information SC is information in which a camera ID, a shooting area ID, and shooting time (shooting time information) information are associated with each other. The shooting time is information indicating the time when the cameras A and B take pictures. The shooting schedule information SC, like the detection program determination table TB, is created by, for example, a user of the image processing device 1 operating the terminal device TM or the input unit 21. The created shooting schedule information SC is stored in the memory unit 24.

[0043] Fig. 4 is a diagram showing an example of photography schedule information in one embodiment of the present invention. In the example shown in Fig. 4, the camera IDs of cameras A and B and the photography area IDs of photography areas A1, A2, B1 to B3 are shown in the same way as the example shown in Fig. 3. In addition, in the example shown in Fig. 4, the photography time is shown as a specific time every day (xx:xx minutes every day) or a specific time every hour (xx minutes every hour). Note that it is also possible to set the photography time as a time specified by, for example, year, month, date, hour, and minute.

[0044] In the shooting schedule information SC shown in Fig. 4, "AAAAA", "A-1", and "5:10 every day" are associated with each other, and "AAAAA", "A-2", and "30 minutes past every hour" are associated with each other. Also, "BBBBB", "B-1", and "7:00 every day" are associated with each other, "BBBBBB", "B-2", and "8:00 every day" are associated with each other, and "BBBBB", "B-3", and "00 minutes past every hour" are associated with each other.

[0045] The processing unit 25 includes a control unit 31, a detection unit 32, an evaluation unit 33, and a report creation unit 34. The processing unit 25 uses images acquired from the cameras A and B to perform various processes required for monitoring.

[0046] The control unit 31 instructs the cameras A and B to set the shooting area to a predetermined shooting area. For example, the control unit 31 gives the above instruction based on the shooting schedule information SC stored in the storage unit 24. The control unit 31 can also set the shooting schedule information SC stored in the storage unit 24 to the cameras A and B. When the shooting schedule information SC is set to the cameras A and B, the cameras A and B will perform shooting based on the shooting schedule information SC even if the control unit 31 does not give the above instruction.

[0047] The detection unit 32 uses a detection program determination table TB stored in the storage unit 24 to identify a detection program PG corresponding to the image capturing areas of the cameras A and B. The detection unit 32 uses the identified detection program PG to detect instruments from the images captured by the cameras A and B, and performs processing to obtain the indicated values ​​of the detected instruments.

[0048] Here, when the control unit 31 issues the above-mentioned instructions to the cameras A and B, the detection unit 32 identifies the detection program PG using the detection program determination table TB based on the shooting area ID and camera ID that identify the shooting area set by the control unit 31. On the other hand, when the control unit 31 sets the shooting schedule information SC to the cameras A and B, the detection unit 32 identifies the detection program PG using the detection program determination table TB based on the camera ID and shooting area ID transmitted from the cameras A and B when the cameras A and B set the shooting area to a predetermined shooting area.

[0049] The evaluation unit 33 evaluates the detection result of the detection unit 32. For example, the evaluation unit 33 evaluates whether the indicated value of the meter obtained by the detection unit 32 exceeds a predetermined threshold value. The evaluation unit 33 outputs the evaluation result to the report creation unit 34. The report creation unit 34 creates a report to report the monitoring result. The report creation unit 34 includes the evaluation result of the evaluation unit 33 in the report, or changes the content of the report depending on the evaluation result of the evaluation unit 33.

[0050] The communication control unit 11 shown in Fig. 1 corresponds to, for example, the communication unit 23 and the control unit 31 shown in Fig. 2. The signal processing unit 12 shown in Fig. 1 corresponds to, for example, the detection unit 32, the evaluation unit 33, and the report creation unit 34 shown in Fig. 2. The output unit 13 shown in Fig. 1 corresponds to, for example, the display unit 22 and the communication unit 23 shown in Fig. 2.

[0051] The functions of the processing unit 25 of the image processing device 1 (the functions of the control unit 31, the detection unit 32, the evaluation unit 33, and the report creation unit 34) are realized by executing a program that realizes these functions on hardware such as a CPU (Central Processing Unit). In other words, the functions of the processing unit 25 of the image processing device 1 are realized by software and hardware resources working together. By realizing the functions of the processing unit 25 of the image processing device 1 through the cooperation of software and hardware resources, it is possible to extremely easily update, delete, add, etc. functions, for example.

[0052] Although it is desirable that the functions of the image processing device 1 be realized by a combination of software and hardware resources, this does not preclude the use of dedicated hardware. The functions of the image processing device 1 may also be realized by using hardware such as a field-programmable gate array (FPGA), large-scale integration (LSI), or application-specific integrated circuit (ASIC).

[0053] <Operation of the monitoring system> Next, the operation of the monitoring system will be described. The operation of the monitoring system SY differs slightly depending on whether the shooting schedule information SC is not set for the cameras A and B or whether the shooting schedule information SC is set for the cameras A and B. When the shooting schedule information SC is not set for the cameras A and B, the image processing device 1 instructs the cameras A and B on the shooting area based on the shooting schedule information SC and causes them to shoot. On the other hand, when the shooting schedule information SC is set for the cameras A and B, the cameras A and B set the shooting area based on the shooting schedule information SC and shoot.

[0054] Fig. 5 is a flowchart showing an example of the operation of an image processing device according to one embodiment of the present invention. Specifically, the flowchart shown in Fig. 5 is a flowchart for the case where the image processing device 1 instructs the cameras A and B on the shooting area based on the shooting schedule information SC and causes them to take pictures. The flowchart shown in Fig. 5 is performed, for example, every time the shooting time set in the shooting schedule information SC arrives.

[0055] 5 starts, first, the control unit 31 of the image processing device 1 instructs the cameras A and B to set a shooting area (step S11). Specifically, the control unit 31 reads out from the shooting schedule information SC the camera ID and shooting area ID associated with the shooting time that matches the current time among the shooting times set in the shooting schedule information SC. Then, the control unit 31 instructs the camera identified by the read camera ID to set its shooting area to the shooting area identified by the read shooting area ID.

[0056] For example, if the current time is 5:10, the control unit 31 reads out "AAAAA" and "A-1" which are associated with "everyday 5:10" shown in Fig. 4. The control unit 31 then instructs the camera A identified by the read-out "AAAAA" to set the shooting area of ​​the camera A to the shooting area identified by "A-1". When such an instruction is given, the camera A sets the shooting area to the shooting area identified by "A-1" based on the instruction from the control unit 31, and takes pictures of the set shooting area.

[0057] Next, the detection unit 32 of the image processing device 1 uses the detection program determination table TB to identify the detection program PG based on the instruction content of the control unit 31 (step S12). Specifically, the detection unit 32 identifies the detection program PG by reading out, from the detection program determination table TB, the detection program ID associated with the camera ID and the shooting area ID read out by the control unit 31.

[0058] For example, suppose the camera ID read by the control unit 31 is "AAAAA" and the shooting area ID is "A-1." The detection unit 32 reads "α," which is associated with "AAAAA" and "A-1," from the detection program determination table TB shown in Fig. 3. This identifies the detection program PG as a detection program for a "circular one-hand meter."

[0059] Next, processing unit 25 controls communication unit 23 to acquire the image captured by the camera (step S13). For example, if control unit 31 issues a setting instruction to camera A in step S11, processing unit 25 causes the image captured by camera A to be acquired.

[0060] Next, the detection unit 32 performs a detection process using the detection program identified in step S12 (step S14: second step). Specifically, the detection unit 32 starts the detection program PG identified in step S12 and inputs the image acquired in step S13. As a result, for example, if the detection program PG is a detection program for a "circular one-needle meter," the "circular one-needle meter" included in the image acquired in step S13 is detected, and its indicated value is acquired.

[0061] Next, the evaluation unit 33 evaluates the detection result obtained in step S14 (step S15). For example, the evaluation unit 33 evaluates whether the indicated value of the "circular one-needle meter" obtained in step S14 exceeds a predetermined threshold value.

[0062] When the above processing is completed, the report creation unit 34 creates a report (a report on the monitoring results) (step S16). Specifically, the report creation unit 34 creates the report based on the image acquired in step S13, the detection results in step S14, the evaluation results in step S15, etc. The format of the report is not particularly limited and can be any format.

[0063] For example, the report creation unit 34 creates a report that includes an image of a "circular one-needle meter" and indicates its indicated value together with a threshold value. The report creation unit 34 may also create a report that includes the evaluation result of step S15. For example, if the evaluation result shows that the indicated value of the meter is above or below a threshold value, the confirmation result in the report may be set to "abnormal" or a comment corresponding to this may be automatically added to the report content.

[0064] 5, for ease of understanding, the process of step S16 is performed after step S15. However, the process of step S16 may be performed after the process of step S15 has been performed a predetermined number of times. In this way, a report showing the monitoring results over a certain period of time can be created.

[0065] The report created by the report creation unit 34 can be displayed on the display unit 22, for example, by a user of the image processing device 1 operating the input unit 21 to give a predetermined instruction. Alternatively, the user of the image processing device 1 can operate the terminal device TM to make a request to the image processing device 1, and the report can be output as data from the communication unit 23 to the terminal device TM. Each time a report is created, the created report may be uploaded to a specific server set in advance, or sent by email or the like to a specific terminal device (e.g., a smartphone or tablet). The report uploaded to the server may be made accessible using a terminal device such as a smartphone or tablet.

[0066] Fig. 6 is a flowchart showing another example of the operation of an image processing device according to one embodiment of the present invention. Specifically, the flowchart shown in Fig. 6 is a flowchart in which cameras A and B set a shooting area and take pictures based on shooting schedule information SC. In Fig. 6, steps that are the same as those shown in Fig. 5 are given the same reference numerals.

[0067] When the shooting time set in the shooting schedule information SC arrives, the cameras A and B read out the shooting area ID from the shooting schedule information SC. Specifically, the cameras A and B read out the shooting area ID whose camera ID is associated with their own camera ID from among the camera IDs and shooting area IDs associated with the shooting time that matches the current time.

[0068] Cameras A and B set the shooting area to the shooting area identified by the read shooting area ID and capture an image of the set shooting area. Then, they transmit setting information indicating that the shooting area has been set to the image processing device 1. The setting information transmitted from cameras A and B to the image processing device 1 includes a shooting area ID that identifies the shooting area after setting. Note that the setting information may also include a camera ID. The processing of the flowchart shown in FIG. 6 is performed each time the communication unit 23 of the image processing device 1 receives setting information transmitted from cameras A and B.

[0069] 6 starts, first, the processing unit 25 of the image processing device 1 acquires the setting information transmitted from the cameras A and B and received by the communication unit 23 (step S21). Next, the detection unit 32 of the image processing device 1 identifies the detection program PG based on the acquired setting information using the detection program determination table TB (step S22). Specifically, the detection unit 32 identifies the detection program PG by reading, from the detection program determination table TB, the detection program ID associated with the camera ID and shooting area ID included in the setting information.

[0070] For example, suppose the camera ID included in the setting information is "AAAAA" and the shooting area ID is "A-2." The detection unit 32 reads out "β," which is associated with "AAAAA" and "A-2," from the detection program determination table TB shown in FIG. 3. This identifies the detection program PG as a detection program for a "circular two-hand meter."

[0071] Once the detection program PG is identified, the following steps are performed, similar to the flowchart shown in Fig. 5: acquiring an image captured by a camera (step S13), performing detection using the identified detection program (step S14), evaluating the detection results (step S15), and creating a report (step S16). Note that the processes of steps S13 to S16 shown in Fig. 6 are the same as those of steps S13 to S16 shown in Fig. 5, and therefore will not be described in detail.

[0072] As described above, in this embodiment, a detection program determination table TB in which camera IDs, photographing area IDs, and detection program IDs are associated with each other is stored in the storage unit 24. In this embodiment, the detection program determination table TB is used to identify a detection program PG corresponding to the photographing areas A1, A2, B1 to B3 of cameras A and B, the identified detection program PG is used to detect instruments from images photographed by cameras A and B, and the indicated values ​​of the detected instruments are obtained. As a result, for each of the photographing areas A1, A2, B1 to B3 of cameras A and B, instruments are detected using an appropriate detection program PG corresponding to the photographing areas A1, A2, B1 to B3 of cameras A and B, so that even if instruments are dispersed, it is possible to detect the instruments efficiently and accurately with a relatively small number of cameras.

[0073] <Variations> (1) Before performing the process of detecting instruments from the images captured by the cameras A and B using the identified detection program PG, the detection unit 32 may perform preprocessing on the images captured by the cameras A and B. Examples of such preprocessing include correction of image distortion caused by the lens characteristics of the cameras A and B, and correction of image distortion caused by the positional relationship between the cameras A and B and the instruments.

[0074] For example, if the lens characteristics of cameras A and B change depending on the zoom level, causing distortion in the captured image, the captured image is subjected to distortion correction as preprocessing. Also, if the instrument is not directly facing cameras A and B, causing distortion in the captured image, the captured image is subjected to trapezoidal distortion correction and projective transformation as preprocessing. By performing such preprocessing, the detection accuracy of the detection program can be improved.

[0075] FIG. 7 is a diagram showing an example of a detection program determination table according to a modified example of one embodiment of the present invention. As shown in FIG. 7, the detection program determination table TB1 according to this modified example is a table in which a preprocessing ID (preprocessing identification information) is associated with a camera ID, a shooting area ID, and a detection program ID. In addition to the preprocessing ID, parameters used in each preprocessing may also be associated. The preprocessing ID is information that identifies the preprocessing performed by the detection unit 32. Like the detection program determination table TB, the detection program determination table TB1 is created, for example, by a user of the image processing device 1 operating the terminal device TM or the input unit 21. The created detection program determination table TB1 is stored in the memory unit 24 (first step).

[0076] In the example shown in Fig. 7, the preprocessing ID for "barrel distortion correction" is represented by "Pr1," the preprocessing ID for "pincushion distortion correction" is represented by "Pr2," the preprocessing ID for "keystone distortion correction" is represented by "Pr3," and the preprocessing ID for "projection transformation" is represented by "Pr4."

[0077] 7, the detection unit 32 identifies a detection program PG according to the image capturing areas of the cameras A and B, and also identifies pre-processing according to the image capturing areas of the cameras A and B. The detection unit 32 performs the identified pre-processing on the images captured by the cameras A and B, and uses the identified detection program PG to detect instruments from the pre-processed images and perform processing to obtain the indicated values ​​of the detected instruments.

[0078] Note that pre-processing may be switched for each shooting area without using the detection program determination table TB1 shown in Fig. 7. For example, to correct image distortion caused by the positional relationship between cameras A and B and the instrument, the orientation of the instrument may be estimated from the image, and keystone distortion correction or the like may be performed according to the estimated orientation.

[0079] (2) The control unit 31 may perform control to change the shooting conditions of the cameras A and B according to the images captured by the cameras A and B. Examples of control to change the shooting conditions of the cameras A and B include control to turn on and off the power of a lighting device or an infrared irradiation device corresponding to the shooting area according to the brightness or contrast of the image.

[0080] If the brightness or contrast of the image is low, the detection accuracy of the detection program PG may decrease. Therefore, for example, when photographing an instrument installed in a dark photographing area, the detection accuracy can be improved by improving the photographing conditions by turning on the power of the lighting device or infrared irradiation device corresponding to the photographing area before switching the photographing area. Note that the user of the image processing device 1 may manually turn on and off the power of the lighting device or infrared irradiation device corresponding to the photographing area.

[0081] (3) The control unit 31 may change the photographing time in accordance with the state of the instrument detected by the detection unit 32. Specifically, the control unit 31 may change the photographing time in accordance with the indication value obtained by the detection unit 32 or the evaluation result of the evaluation unit 33. For example, if the fluctuation range of the indication value obtained in the past by the detection unit 32 is large or if the evaluation unit 33 frequently detects abnormalities, the control unit 31 changes the photographing time so as to shorten the photographing cycle.

[0082] The image processing device, image processing method, and image processing program according to the embodiments and modifications of the present invention have been described above. However, the present invention is not limited to the above embodiments and modifications and can be freely modified within the scope of the present invention. For example, in the above-described embodiments, the object is an instrument, but the object is not limited to an instrument and may be a person, a vehicle, or the like. If the object is a person or a vehicle, a detection program PG for detecting the person or vehicle must be prepared. Furthermore, if the object is a person, a detection program PG for detecting a specific movement of the person (such as a fall or suspicious behavior) may also be prepared.

[0083] In the above-described embodiment, the shooting areas A1, A2, B1 to B3 of the cameras A and B are set to capture only one instrument. If multiple instruments are captured in the shooting areas A1, A2, B1 to B3 of the cameras A and B, it is also possible to perform detection using a method similar to that used in the past. In other words, it is also possible to detect multiple instruments by using multiple detection programs PG. [Explanation of symbols]

[0084] 1. Image processing device 24 Memory section 31 Control Unit 32 Detection unit Camera A, B A1, A2 photo area B1~B3 Photo area PG detection program SC Shooting Schedule Information TB,TB1 Detection program decision table

Claims

1. a storage unit that stores table information in which at least photographing area identification information that identifies a photographing area of ​​a camera whose photographing area is changeable and detection program identification information that identifies a detection program that detects a predetermined object from an image photographed by the camera are associated with each other; a detection unit that uses the table information stored in the storage unit to identify the detection program corresponding to the photographing area of ​​the camera, and performs processing to detect the object from the image photographed by the camera using the identified detection program; An image processing device comprising:

2. a control unit that instructs the camera to set a photographing area of ​​the camera to a predetermined photographing area; the detection unit specifies the detection program using the table information based on the photography area identification information that identifies the predetermined photography area set by the control unit.

2. The image processing device according to claim 1.

3. the storage unit stores at least photography schedule information in which photography time information specifying the time when photography is performed by the camera is associated with the photography area identification information; The image processing device according to claim 2 , wherein the control unit issues the instruction based on the photographing schedule information stored in the storage unit.

4. 2. The image processing device according to claim 1, wherein the detection unit identifies the detection program using the table information based on the shooting area identification information that identifies the predetermined shooting area transmitted from the camera when the camera sets the shooting area to a predetermined shooting area.

5. 5. The image processing device according to claim 4, further comprising a control unit that causes the camera to set, in the camera, photography schedule information in which photography time information that specifies the time at which photography is to be performed by the camera is associated with the photography area identification information.

6. the table information is information in which camera identification information for identifying the plurality of cameras, the photography area identification information, and the detection program identification information are associated with each other; the photography schedule information is information in which the photography time information, the camera identification information, and the photography area identification information are associated with each other; the detection unit and the control unit identify the camera using the camera identification information; 6. The image processing device according to claim 3 or claim 5.

7. The image processing device according to claim 1 , wherein the detection unit performs preprocessing on the image captured by the camera before performing processing to detect the object from the image captured by the camera using the identified detection program.

8. the table information is information in which the photography area identification information, the detection program identification information, and pre-processing identification information for identifying the pre-processing are associated with each other, the detection unit performs the pre-processing according to the photographing area of ​​the camera using the table information.

8. The image processing device according to claim 7.

9. 6. The image processing device according to claim 3, wherein the control unit controls the camera to change a photographing condition in accordance with an image photographed by the camera.

10. 6. The image processing device according to claim 3, wherein the control unit changes the photographing time information in accordance with the state of the object detected by the detection unit.

11. a first step in which a storage unit stores table information in which at least photographing area identification information for identifying a photographing area of ​​a camera whose photographing area is changeable and detection program identification information for identifying a detection program for detecting a predetermined object from an image photographed by the camera are associated with each other; a second step in which a detection unit uses the table information stored in the storage unit to identify the detection program corresponding to the photographing area of ​​the camera, and performs processing to detect the object from the image photographed by the camera using the identified detection program; An image processing method comprising:

12. On the computer, a first step of storing table information in which at least photographing area identification information for identifying a photographing area of ​​a camera whose photographing area is changeable and detection program identification information for identifying a detection program for detecting a predetermined object from an image photographed by the camera are associated with each other; a second step of identifying the detection program according to the photographing area of ​​the camera using the table information, and performing processing to detect the object from the image photographed by the camera using the identified detection program; An image processing program for executing the above.

Images