Lighting confirmation system and lighting confirmation method

Abstract

To provide a lighting confirmation system and a lighting confirmation method capable of easily confirming the lighting state of a lighting lamp. [Solution] The system comprises an image / position acquisition means 10 that acquires an image taken in the direction of travel of the moving body by an imaging means mounted on the moving body and the image position of the image, an illumination lamp position registration means 20 that registers the position of the illumination lamp, a target illumination lamp extraction means 30 that extracts, from the illumination lamps, those located within a predetermined extraction range in the direction of travel from the imaging position as target illumination lamps, and a determination means 40 that determines whether the light from the target illumination lamp is captured in a corresponding area of ​​the captured image, taking advantage of the fact that the area in the captured image in which the light from the target illumination lamp is captured differs depending on the distance between the imaging position and the position of the target illumination lamp.

Description

[Technical Field]

[0001] The present invention relates to a lighting confirmation system and a lighting confirmation method for confirming the lighting status of a plurality of lighting lamps arranged along a road. [Background technology]

[0002] On expressways and other roads, road lighting equipment is inspected periodically to ensure safe nighttime travel. Conventionally, maintenance personnel would patrol the roads at night and conduct visual inspections. However, since inspections are performed at night, this places a heavy burden on maintenance personnel and poses the risk of accidents due to vehicle patrols. Furthermore, since vehicles emit exhaust gases, improvements are also needed from the perspective of environmental conservation.

[0003] In recent years, it has also been proposed to use patrol cars to conduct lighting inspections. For example, Patent Document 1 describes a device equipped with a lighting detector on a vehicle that detects the lighting state based on the illumination light from a lighting lamp installed along a road when the vehicle is near the lighting detector. This lighting detector has two cameras spaced apart so that they are parallel to each other and facing the same direction, and the cameras are tilted at a predetermined angle between 15° and 25° from the vertical to the left of the vehicle's direction of travel in a plane perpendicular to the vehicle's direction of travel. This lighting detector distinguishes between lighting lamps and noise light and detects the illumination light from lighting lamps, since noise light from building lights, signboard lights, etc. is located farther from the cameras than lighting lamps. This lighting detector utilizes the fact that the horizontal positional deviation of the lighting lamp image is larger than that of the noise light image in the original images from the two cameras.

[0004] However, the device in Patent Document 1 requires two cameras because it utilizes the fact that the amount of horizontal shift that occurs in the images from the two cameras differs depending on the distance between the cameras and the object, and it also requires detailed adjustments to the camera installation angle, field of view, aperture, etc., resulting in a complex structure. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] Japanese Patent Application Laid-Open No. 2001-85174 Summary of the Invention [Problem to be solved by the invention]

[0006] The present invention has been made in view of the above-mentioned problems, and has as its object to provide a lighting confirmation system and a lighting confirmation method that can easily confirm the lighting state of lighting lamps. [Means for solving the problem]

[0007] The lighting confirmation system of the present invention is for confirming the lighting status of a plurality of lighting lamps arranged along a road, and comprises: a photographed image and position acquisition means for acquiring a photographed image taken in the direction of travel of a mobile body by a photographing means mounted on the mobile body moving along the road, and the photographed position of the photographed image; a lighting lamp position registration means for registering the positions of the plurality of lighting lamps; a target lighting lamp extraction means for extracting, from the lighting lamps registered in the lighting lamp position registration means, a lighting lamp that is located within a predetermined extraction range in the direction of travel from the photographed position as a target lighting lamp; and a determination means for determining whether or not the light of the target lighting lamp is captured in a corresponding area of ​​the photographed image, by utilizing the fact that the area in which the light of the target lighting lamp is captured differs depending on the distance between the photographed position and the position of the target lighting lamp.

[0008] The lighting confirmation method of the present invention is for confirming the lighting status of a plurality of lighting lamps arranged along a road, and includes a photographed image and position acquisition step for acquiring a photographed image taken in the traveling direction of a moving body by a photographing means mounted on the moving body moving along the road and the photographing position of the photographed image; a lighting lamp position registration step for registering the positions of the plurality of lighting lamps; a target lighting lamp extraction step for extracting, from the lighting lamps registered in the lighting lamp position registration step, lighting lamps located within a predetermined extraction range in the traveling direction from the photographing position as target lighting lamps; and a determination step for determining whether or not light from the target lighting lamp is captured in a corresponding area of ​​the photographed image, by utilizing the fact that the area in which light from the target lighting lamp is captured differs depending on the distance between the photographing position and the position of the target lighting lamp. [Effects of the Invention]

[0009] According to the present invention, lighting fixtures located within a predetermined extraction range in the direction of travel from the shooting position are extracted as target lighting fixtures, and whether or not the light from the target lighting fixture is captured in the corresponding area of ​​the captured image is determined by utilizing the fact that the area in the captured image in which the light from the target lighting fixture is captured differs depending on the distance between the shooting position and the position of the target lighting fixture, so that the lighting status of the lighting fixture can be confirmed easily and with high accuracy.

[0010] Furthermore, by dividing the captured image into a first region corresponding to a first range within the extraction range that is farther than a predetermined reference distance from the shooting position, and a second region corresponding to a second range within the extraction range that is from the shooting position to the reference distance, and determining whether or not the light from the target lamp is captured, the lighting status can be confirmed with higher accuracy.

[0011] Furthermore, by extracting light candidates for the target lighting lamp from the binary image of the captured image and determining from the light candidates that the light is larger than a predetermined reference size as the light of the target lighting lamp, the lighting state can be confirmed with higher accuracy. [Brief explanation of the drawings]

[0012] [Figure 1]1 is a diagram illustrating a configuration of a lighting lighting confirmation system according to an embodiment of the present invention. [Figure 2] FIG. 10 is a diagram illustrating an example of the positions of illuminating lamps, the photographing positions, and the extraction range on an electronic map. [Figure 3] FIG. 2 is a diagram illustrating an example of a binarized image of a captured image. [Figure 4] FIG. 10 is a diagram showing an example of a first range and a second range in an extraction range. [Figure 5] FIG. 2 is a diagram showing an example of a first region and a second region in a captured image. [Figure 6] 1 is a diagram illustrating a flow of a method for checking whether a light is on according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.

[0014] 1 shows the configuration of a lighting confirmation system 10 according to one embodiment of the present invention. This lighting confirmation system 1 is used to confirm the lighting status of a plurality of lighting lamps arranged along a road. Specifically, for example, it is used to confirm whether a plurality of lighting lamps arranged at intervals along a road such as a highway are turned on.

[0015] This lighting confirmation system 1 includes, for example, an image capturing means and position capturing means 10 for capturing an image of the moving body's direction of travel by an image capturing means mounted on the moving body as it moves along a road, and for capturing the image capturing position; an illumination lamp position registering means 20 for registering the positions of a plurality of illumination lamps arranged along the road; a target illumination lamp extracting means 30 for extracting a target illumination lamp from the illumination lamps registered in the illumination lamp position registering means 20; a determination means 40 for determining whether or not the light from the target illumination lamp is captured in the captured image; and a determination result recording means 50 for recording the determination result.

[0016] The photographed image and position acquisition means 10, the lamp position registration means 20, the target lamp extraction means 30, the judgment means 40, and the judgment result recording means 50 can be configured, for example, by a computer, and is configured to function as the photographed image and position acquisition means 10, the lamp position registration means 20, the target lamp extraction means 30, the judgment means 40, or the judgment result recording means 50 by executing a program.

[0017] The captured image and position acquisition means 10 acquires, for example, captured images taken by a camera mounted on a mobile object moving along a road, in association with the shooting position at the time the captured image was taken. The captured images and shooting positions can be acquired by receiving them via a network such as the Internet. Alternatively, the captured images and shooting positions may be saved in a storage means such as a memory, and the storage means may be connected to the captured image and position acquisition means 10 and read from the storage means to acquire the captured images and shooting positions. The acquisition of the captured images and shooting positions may be performed by real-time processing, in which data is processed as needed, or by batch processing, in which data is processed at regular intervals.

[0018] A preferred example of a mobile body equipped with an imaging device is a patrol car that periodically patrols roads. An example of the imaging device is a CCD camera, which is installed so as to capture images in the direction of travel of the mobile body. The imaging position can be obtained, for example, by using a GPS (Global Positioning System). For example, it is preferable to use a mobile terminal such as a smartphone equipped with a CCD camera and a GPS receiver as the imaging device, as this makes it easy to obtain the captured image and the imaging position.

[0019] The captured images may be continuous video or still images taken at predetermined intervals, for example, every 5 seconds. However, it is preferable to take still images at predetermined intervals. This is because it reduces the burden of transmitting and storing the captured image data and also makes it possible to input coordinates for identifying the facility location as property information. It is preferable that the captured images be taken at night.

[0020] The lamp position registration means 20 is used to register, for example, the positions of lamps in advance. The positions of lamps may be, for example, coordinates (latitude and longitude) where the lamps are installed. The positions of lamps can be obtained, for example, by writing the position coordinates into a management drawing file and editing and locating the file. Furthermore, the lamp position registration means 20 is preferably configured to be able to display the registered positions of lamps on an electronic map. FIG. 2 shows an example in which the positions of lamps are represented by black map pins on an electronic map. Note that the areas shown in gray in FIG. 2 are roads.

[0021] The target luminaire extraction means 30 extracts, from among the luminaires registered in the luminaire position registration means 20, those luminaires located within a predetermined extraction range in the traveling direction from the photographing position as target luminaires. In FIG. 2, the position indicated by the white map pin is the photographing position, and the area surrounded by the dashed line is the extraction range. The extraction range is, for example, a range from the photographing position to a position a predetermined distance away in the traveling direction of the moving object. In other words, it is preferable to extract, as target luminaires, luminaires whose positions are within a predetermined distance from the photographing position in the traveling direction of the photographing position. This is because by extracting luminaires whose positions are within a predetermined distance from the photographing position, the lighting status can be determined with high accuracy. The traveling direction of the moving object can be determined from the photographing position, i.e., the moving direction of the photographing position coordinates.

[0022] The extraction range is preferably set, for example, within a range from 20 m away from the shooting position to 120 m away from the shooting position. In other words, it is preferable to set the extraction range so that the maximum distance between the shooting position and the position of the illuminating lamp is, for example, between 20 m and 120 m. The target illuminating lamp can be extracted, for example, by searching the position of the illuminating lamp registered in the illuminating lamp position registration means 20, i.e., by searching the coordinates.

[0023] The determination means 40 determines whether or not the light from the target lamp is captured in a corresponding area of ​​the photographed image, utilizing the fact that the area in the photographed image in which the light from the target lamp is captured differs depending on the distance between the photographing position and the position of the target lamp. This is because when the distance between the photographing position and the position of the target lamp is far, the light from the target lamp is captured in an area in the center of the photographed image, and when the distance between the photographing position and the position of the target lamp is close, the light from the target lamp is captured in areas on the upper side and both the left and right sides of the photographed image, thereby determining whether or not the light from the target lamp is captured in the photographed image.

[0024] The determination means 40 preferably includes, for example, binarization processing means 41 that binarizes the captured image using a predetermined threshold to obtain a binary image, and light candidate extraction means 42 that extracts light candidates for the target lighting fixture from the binary image obtained by the binarization processing means 41. The binarization processing means 41 simplifies the image to make it easier to analyze. The threshold is preferably adjusted depending on the ambient brightness, for example, the season, sunrise time, sunset time, etc. Figure 3 shows an example of a binary image.

[0025] The light candidate extraction means 42 is preferably configured to extract, as light candidates, portions that are thought to be light from target lighting fixtures from a binary image by, for example, image recognition using AI (Artificial Intelligence). The light candidate extraction means 42 preferably includes, for example, a light candidate learning model 42A that has learned the characteristics of lighting fixture light by machine learning, a detection means 42B that detects light candidates from the binary image by image recognition using the light candidate learning model 42A, and a light candidate learning model generation means 42C that generates the light candidate learning model 42A.

[0026] The light candidate extraction means 42 preferably also has a size determination means 42D that extracts, from the light candidates detected by the detection means 42B, those that are larger than a predetermined reference size as light from the target illuminating lamp, and a target light recording means 42E that records the position and size of the light from the extracted target illuminating lamp.

[0027] Furthermore, it is preferable that the determination means 40 has, for example, a distance calculation means 43 that calculates the distance between the position of the target lamp extracted by the target lamp extraction means 30 and the photographing position, and a matching means 44 that, based on the calculation result of the distance calculation means 43, matches and determines whether or not light from the target lamp recorded in the target light recording means 42E is present in the corresponding area of ​​the photographed image.

[0028] The matching means 44 is preferably configured to determine whether or not light from a target lamp is captured by dividing the captured image into a first region corresponding to a first region within the extraction range that is farther from the shooting position than a predetermined reference distance, and a second region corresponding to a second region within the extraction range that is a predetermined reference distance from the shooting position. In FIG. 4, the region surrounded by a dashed line is the extraction range, the region surrounded by a dotted line is the first region, and the region surrounded by a dashed-dotted line is the second region. In FIG. 5, the region surrounded by a dotted line is the first region, and the region surrounded by a dashed-dotted line is the second region. Light from a target lamp located in the first region far from the shooting position within the extraction range is captured in the first region in the center of the captured image, and light from a target lamp located in the second region within the extraction range that is closer to the shooting position is captured in the second region on the upper and left and right sides of the captured image. Therefore, by selecting and matching the first region or the second region depending on the distance between the shooting position and the position of the target lamp, highly accurate determination can be made. In Figure 4, the areas shown in gray are roads, the locations indicated by black map pins are the locations of lighting fixtures, and the locations indicated by white map pins are the shooting locations.

[0029] The matching means 44 is preferably configured to, for example, determine that the target illuminant is on if there is light from the target illuminant in the first area when the target illuminant is located in the first range, and determine that the target illuminant is off if there is no light from the target illuminant in the first area when the target illuminant is located in the second range, and determine that the target illuminant is on if there is light from the target illuminant in the second area when there is no light from the target illuminant in the second area when the target illuminant is located in the second range. The reference distance separating the first and second ranges can be set as appropriate, and is preferably set within the range of, for example, 20 to 50 meters.

[0030] This illumination lighting confirmation system 1 checks the lighting state of a plurality of illumination lamps arranged along a road, for example, as follows: Figure 6 shows the flow of the illumination lighting confirmation method.

[0031] In this method for checking whether a light is on, first, the captured image and position acquisition means 10 acquires a captured image of the moving body's direction of travel by an image capture means mounted on the moving body moving along a road, and the capture position of the captured image (step S110; captured image and position acquisition procedure).Then, the positions of a plurality of lights arranged along the road are registered by the light position registration means 20 (step S120; light position registration procedure).

[0032] Next, the target illuminant extraction means 30 extracts, from among the illuminants registered in the illuminant position registration procedure (step S120), those located within a predetermined extraction range in the traveling direction from the photographing position as target illuminants (step S130; target illuminant extraction procedure). In other words, it is preferable to extract, as target illuminants, illuminants whose positions are within a predetermined distance from the photographing position in the traveling direction from the photographing position. The extraction range is as described above.

[0033] Next, the determination means 40 determines whether the light from the target lamp is captured in the corresponding area of ​​the captured image, utilizing the fact that the area in the captured image in which the light from the target lamp is captured differs depending on the distance between the shooting position and the position of the target lamp (step S140; determination procedure).

[0034] In the determination procedure (step S140), for example, first, the binarization processing means 41 binarizes the captured image using a predetermined threshold to generate a binary image (step S141; binarization processing procedure). Next, the light candidate extraction means 42 extracts light candidates of the target illuminating lamp from the binary image (step S142; light candidate extraction procedure). For example, it is preferable to detect light candidates of the target illuminating lamp by image recognition using AI, specifically, by using a light candidate learning model 42A that has learned the characteristics of the light of the illuminating lamp by machine learning, and then extract from the light candidates those that are larger than a predetermined reference size by the size determination means 42D as the light of the target illuminating lamp.

[0035] In the determination procedure (step S140), the distance between the position of the target illuminant and the photographing position is calculated by the distance calculation means 43 (step S143; distance calculation procedure). Next, based on the calculation result, the matching means 44 performs a match to determine whether or not light from the target illuminant is present in the corresponding area of ​​the photographed image (step S144; match procedure).

[0036] In this case, it is preferable to divide the captured image into a first region corresponding to a first range within the extraction range that is farther than a predetermined reference distance from the shooting position, and a second region corresponding to a second range within the extraction range that is up to the predetermined reference distance from the shooting position, and compare these regions to determine whether or not there is light from the target illuminant. Specifically, for example, when the target illuminant is located in the first range, if there is light from the target illuminant in the first region, it is determined to be on, and if there is no light from the target illuminant, it is determined to be off. When the target illuminant is located in the second range, if there is light from the target illuminant in the second region, it is determined to be on, and if there is no light from the target illuminant, it is determined to be off. The determination result is recorded, for example, by the determination result recording means 50 (step S150; determination result recording procedure).

[0037] As described above, according to this embodiment, lamps located within a predetermined extraction range in the traveling direction from the shooting position are extracted as target lamps, and whether or not the light from the target lamp is captured in the corresponding area of ​​the captured image is determined by utilizing the fact that the area in the captured image in which the light from the target lamp is captured differs depending on the distance between the shooting position and the position of the target lamp. Therefore, the lighting status of the lamps can be checked easily and with high accuracy.

[0038] Furthermore, by dividing the captured image into a first region corresponding to a first range within the extraction range that is farther than a predetermined reference distance from the shooting position, and a second region corresponding to a second range within the extraction range that is from the shooting position to the reference distance, and determining whether or not the light from the target lamp is captured, the lighting status can be confirmed with higher accuracy.

[0039] Furthermore, by extracting light candidates for the target lighting lamp from the binary image of the captured image and determining from the light candidates that the light is larger than a predetermined reference size as the light of the target lighting lamp, the lighting state can be confirmed with higher accuracy.

[0040] Although the present invention has been described above with reference to the embodiments, the present invention is not limited to the above embodiments and can be modified in various ways. For example, although each component has been specifically described in the above embodiments, not all components may be included, and other components may be included. [Explanation of symbols]

[0041] 1...lighting on confirmation system, 10...captured image / position acquisition means, 20...lighting lamp position registration means, 30...target lighting lamp extraction means, 40...determination means, 41...binarization processing means, 42...light candidate extraction means, 42A...light candidate learning model, 42B...detection means, 42C...light candidate learning model generation means, 42D...size determination means, 42E...target light recording means, 43...distance calculation means, 44...matching means, 50...determination result recording means

Claims

1. A lighting confirmation system that confirms the lighting status of a plurality of lighting lamps arranged along a road, a photographed image and position acquisition means for acquiring a photographed image of a moving body moving along the road in a traveling direction of the moving body by a photographing means mounted on the moving body, and a photographed image and position acquisition means for acquiring a photographed position of the photographed image; a lighting position registration means for registering the positions of the plurality of lightings; a target lamp extraction means for extracting, from among the lamps registered in the lamp position registration means, lamps that are located within a predetermined extraction range in the traveling direction from the photographing position as target lamps; a determining means for determining whether or not light from the target lamp is captured in a corresponding area of ​​the photographed image, utilizing the fact that an area in the photographed image in which light from the target lamp is captured differs depending on the distance between the photographing position and the position of the target lamp; The determination means divides the captured image into a first region corresponding to a first range within the extraction range that is farther than a predetermined reference distance from the shooting position, and a second region corresponding to a second range within the extraction range that is from the shooting position to the reference distance, and determines whether light from the target lamp is captured in the captured image. A lighting confirmation system characterized by the above.

2. The determining means determines whether or not light from the target lamp is captured within a corresponding area based on a binarized image obtained by binarizing the captured image using a predetermined threshold value.

2. The lighting confirmation system according to claim 1.

3. The determining means extracts light candidates for the target illuminating light from the binarized image, and determines whether or not light from the target illuminating light is captured in a corresponding region by determining from the light candidates that a size larger than a predetermined reference size is the light from the target illuminating light.

3. The lighting confirmation system according to claim 2.

4. A lighting confirmation method for confirming the lighting status of a plurality of lighting lamps arranged along a road, comprising: a photographed image and position acquisition step of acquiring a photographed image of a moving body moving along the road in a traveling direction of the moving body by a photographing means mounted on the moving body, and a photographed position of the photographed image; a lighting position registration step of registering the positions of the plurality of lightings; a target lamp extraction step of extracting, from among the lamps registered in the lamp position registration step, lamps that are located within a predetermined extraction range in a traveling direction from the photographing position as target lamps; a determination step of determining whether or not light from the target lamp is captured in a corresponding area of ​​the photographed image, utilizing the fact that an area in the photographed image in which light from the target lamp is captured differs depending on the distance between the photographing position and the position of the target lamp; In the determination step, the captured image is divided into a first region corresponding to a first range within the extraction range that is farther than a predetermined reference distance from the shooting position, and a second region corresponding to a second range within the extraction range that is from the shooting position to the reference distance, and it is determined whether or not light from the target lamp is captured in the captured image. A method for checking whether a light is on or off.

Images