Foreign matter detection system, foreign matter detection device, foreign matter detection method, and foreign matter detection program

Abstract

To highly accurately detect foreign matters left over on a paved road.SOLUTION: First point group acquisition means 101 is configured to acquire a first point group generated by range-finding a paved road at first time. Second point group acquisition means 102 is configured to acquire a second point group generated by range-finding the paved road at second time later than at the first time. First calculation means 103 is configured to calculate a first point density characteristic in a first detection space on the basis of the first point group. Second calculation means 104 is configured to calculate a second point density characteristic in a second detection space corresponding to the first detection space on the basis of the second point group. Foreign matter determination means 105 is configured to determine whether the foreign matter is left over on the paved road at the second time on the basis of the first point density characteristic and second point density characteristic.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a foreign object detection system, a foreign object detection device, a foreign object detection method, and a foreign object detection program. [Background technology]

[0002] When operating aircraft facilities such as airports, it is important to thoroughly inspect the paved surfaces for aircraft, i.e., runways, taxiways, and parking areas. This is because foreign objects may be left behind on the paved surfaces. Foreign objects may be relatively large, such as animal carcasses, relatively small, such as tools used to maintain aircraft, or even minute, such as parts that have fallen off aircraft or patrol vehicles. These foreign objects may cause aircraft tires to burst or be sucked into jet engines, causing fatal damage to the jet engines. To avoid such situations, airport staff must regularly patrol the paved surfaces and make efforts to find and remove foreign objects. However, the paved surfaces are extremely large, and the regular patrols mentioned above require a great deal of expense.

[0003] Patent Document 1 discloses a harvester that harvests agricultural crops in a field. The harvester is equipped with a stereo camera. The harvester generates a three-dimensional point cloud by performing stereo matching using a pair of captured images acquired from the stereo camera. The harvester detects obstacles based on the three-dimensional point cloud. Specifically, the harvester divides the processing target space ahead in the traveling direction into a grid shape in the horizontal and vertical directions. In this way, the harvester divides the processing target space into a plurality of cells. Next, the harvester stores the three-dimensional point cloud in the corresponding cell. The harvester then estimates that an obstacle is present in a cell where the point density exceeds a predetermined threshold value. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Patent No. 7195543 Summary of the Invention [Problem to be solved by the invention]

[0005] Meanwhile, the inventors of the present application are considering installing a LiDAR device in an aircraft facility and detecting foreign objects left behind on a paved road surface for aircraft using a three-dimensional point cloud output from the LiDAR device. Specifically, the inventors of the present application focus on the point density of the three-dimensional point cloud as in the above-mentioned Patent Document 1 when detecting foreign objects. However, when detecting foreign objects based on the point density of the three-dimensional point cloud, there are several unsolvable problems.

[0006] First, LiDAR devices must be installed close to the ground in accordance with relevant regulations. In contrast, paved roads for aircraft span several kilometers horizontally. Therefore, LiDAR devices irradiate laser light toward the paved road surface at an extremely shallow angle. On the other hand, to give the paved road surface drainage, the paved road surface is generally inclined in an inverted V shape. That is, the paved road surface has a first inclined surface and a second inclined surface with the center line as the middle. The first inclined surface is closer to the LiDAR device than the center line. The second inclined surface is farther from the LiDAR device than the center line. In this case, the point density of the first inclined surface will inevitably be higher than the point density of the second inclined surface. As the point density is thus greatly affected by the slope of the paved road surface itself, foreign object detection using point density cannot avoid overdetection or detection omission.

[0007] Secondly, because the LiDAR device emits laser light several kilometers ahead, the distance measurement accuracy deteriorates due to the divergence of the laser light. The deterioration of the distance measurement accuracy itself leads to the density of the points becoming smaller. As the point density is thus greatly affected by the irradiation distance of the laser light from the LiDAR device, foreign object detection using the point density cannot avoid overdetection or oversight.

[0008] In this way, foreign object detection, which is based on the assumption that the point density on the paved road surface will be uniform if there are no foreign objects on the paved road surface, cannot provide a detection accuracy that can withstand actual operation.

[0009] An object of the present disclosure is to provide a technology for detecting foreign objects left behind on a paved road surface with high accuracy. [Means for solving the problem]

[0010] According to a first aspect of the present disclosure, A first point cloud acquisition means for acquiring a first point cloud generated by measuring the distance of a paved road surface at a first time; a second point cloud acquisition means for acquiring a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time; a first calculation means for calculating a first point density distribution characteristic in a first detection space based on the first point group; a second calculation means for calculating a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point group; a foreign object determination means for determining whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic; Including, A foreign object detection system is provided.

[0011] According to a second aspect of the present disclosure, a first point cloud acquisition means for acquiring a first point cloud generated by measuring the distance to a paved road surface at a first time; a second point cloud acquisition means for acquiring a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time; a first calculation means for calculating a first point density distribution characteristic in a first detection space based on the first point group; a second calculation means for calculating a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point group; a foreign object determination means for determining whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic; Including, A foreign object detection device is provided.

[0012] According to a third aspect of the present disclosure, The computer A first point cloud is obtained by measuring the distance of a paved road surface at a first time; acquiring a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time; Calculating a first point density distribution characteristic in a first detection space based on the first point group; calculating a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point group; determining whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic; A method for detecting foreign objects is provided. Effect of the Invention

[0013] According to the present disclosure, foreign objects left behind on a paved road surface can be detected with high accuracy. [Brief description of the drawings]

[0014] [Figure 1] FIG. 2 is a functional block diagram of the foreign object detection system. [Diagram 2] An aerial view of the airport. [Diagram 3] FIG. 2 is a functional block diagram of the foreign object detection device. [Figure 4] FIG. 11 is a data structure diagram of a first ranging point group. [Diagram 5] 1 is a perspective view of a runway with a first range cloud visualized; [Figure 6] FIG. 11 is a data structure diagram of a second ranging point group. [Figure 7] 1 is a perspective view of a runway with a second range cloud visualized; [Figure 8] FIG. 11 is an explanatory diagram of calculation of point density in the first ranging point cloud. [Figure 9] FIG. 11 is a data structure diagram of a first ranging point group. [Figure 10] FIG. 13 is a data structure diagram of an inspection point group. [Figure 11] 11 is an explanatory diagram of calculation of point density distribution characteristics in the first ranging point cloud. FIG. [Figure 12] 11 is a histogram of point density in a first ranging point cloud. [Figure 13] FIG. 13 is a data structure diagram of an inspection point group. [Figure 14] FIG. 11 is an explanatory diagram of calculation of point density in the second ranging point cloud. [Figure 15] FIG. 11 is a data structure diagram of a second ranging point group. [Figure 16] 11 is an explanatory diagram of calculation of point density distribution characteristics in the second ranging point cloud. FIG. [Figure 17] 13 is a histogram of point density in the second ranging point cloud. [Figure 18] FIG. 13 is a data structure diagram of an inspection point group. [Figure 19] FIG. 13 is a diagram showing an example of an image output. [Figure 20] 4 is a control flow of the foreign object detection device. [Figure 21] FIG. 11 is a data structure diagram of a first ranging point group. [Figure 22] FIG. 11 is a data structure diagram of a second ranging point group. [Figure 23] FIG. 13 is a diagram showing an example of an image output. [Figure 24] 4 is a control flow of the foreign object detection device. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] (Summary of the Disclosure) The foreign object detection system of the present disclosure will be outlined below with reference to the drawings.

[0016] As shown in FIG. 1, the foreign object detection system 100 includes a first point cloud acquisition means 101, a second point cloud acquisition means 102, a first calculation means 103, a second calculation means 104, and a foreign object determination means 105.

[0017] The first point cloud acquisition means 101 acquires a first point cloud generated by measuring the distance to a paved road surface at a first time.

[0018] The second point cloud acquisition means 102 acquires a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time.

[0019] The first calculation means 103 calculates a first point density distribution characteristic in the first detection space based on the first point group.

[0020] The second calculation means 104 calculates a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point group.

[0021] The foreign object determining means 105 determines whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic.

[0022] The foreign object detection system 100 can detect foreign objects left behind on a paved road surface with a high degree of accuracy.

[0023] (First embodiment) Next, a first embodiment of the present disclosure will be described with reference to the drawings.

[0024] Fig. 2 shows an overhead view of an airport 1. As shown in Fig. 2, the airport 1 is mainly provided with a runway 2, a taxiway 3, and an aircraft apron 4. The runway 2, the taxiway 3, and the aircraft apron 4 are specific examples of paved road surfaces.

[0025] The length of the runway 2 is typically between 2000 and 4000 metres. The width of the runway 2 is typically between 30 and 60 metres.

[0026] The airport 1 is provided with a foreign object detection system 10 for monitoring the runway 2, the taxiway 3, and the apron 4. The foreign object detection system 10 detects foreign objects left behind on the runway 2, the taxiway 3, and the apron 4. For ease of explanation, the following describes the foreign object detection system 10 detecting a foreign object left behind on the runway 2. The foreign object detection system 10 includes a plurality of LiDAR devices 11 and a foreign object detection device 12.

[0027] In this embodiment, the multiple LiDAR devices 11 include a LiDAR device 11P and a LiDAR device 11Q. The LiDAR device 11P monitors the first road surface 5P, which is far away from the multiple approach lights 13, of the vast runway 2. The LiDAR device 11Q monitors the second road surface 5Q, which is near the multiple approach lights 13, of the vast runway 2. The first road surface 5P and the second road surface 5Q constitute the runway 2. Hereinafter, the monitoring of the first road surface 5P using the LiDAR device 11P will be mainly described. Meanwhile, the monitoring of the second road surface 5Q using the LiDAR device 11Q will be omitted because it would be redundant. The LiDAR device 11Q may be omitted. The LiDAR device 11P may monitor the entire runway 2.

[0028] In this embodiment, each LiDAR device 11 employs a ToF (Time Of Flight) method as a distance measurement method. However, instead of this, each LiDAR device 11 may employ an FMCW (Frequency Modulated Continuous Wave) method or an AMCW (Amplitude-modulated continuous wave) method as a distance measurement method. The FMCW method and the AMCW method are sometimes called indirect ToF methods.

[0029] The higher the installation height of each LiDAR device 11, the more advantageous it is for monitoring the runway 2. However, the installation height of each LiDAR device 11 is strictly limited by aviation laws. The installation height of each LiDAR device 11 is typically 2 meters to 5 meters.

[0030] Each LiDAR device 11 is typically fixedly installed within the grounds of the airport 1. That is, each LiDAR device 11 is a non-vehicle-mounted type. However, instead of this, each LiDAR device 11 may be a portable type. That is, each LiDAR device 11 may be a vehicle-mounted type. However, each LiDAR device 11 is configured to perform scanning while remaining stationary in space.

[0031] 3, the LiDAR device 11P and the LiDAR device 11Q can bidirectionally communicate with the foreign object detection device 12. The foreign object detection device 12 includes a processor 12a, a memory 12b, a communication interface 12c, and an LCD 12d (Liquid Crystal Display). The processor 12a bidirectionally communicates with the LiDAR device 11P and the LiDAR device 11Q via the communication interface 12c.

[0032] The processor 12a reads out and executes a control program stored in the memory 12b. As a result, the control program causes hardware such as the processor 12a to function as a first point cloud acquisition unit 15, a second point cloud acquisition unit 16, a first calculation unit 17, a second calculation unit 18, a foreign object determination unit 19, and an output unit 20. In this embodiment, the foreign object detection device 12 is realized by a single device. However, instead of this, the foreign object detection device 12 may be realized by distributed processing using a plurality of devices.

[0033] The first point cloud acquisition unit 15 acquires a first point cloud generated by measuring the distance to the runway 2 at a first time. Specifically, the LiDAR device 11P measures the distance to the first road surface 5P of the runway 2 at a first time to generate the first point cloud. The first point cloud is typically a point cloud expressed in an orthogonal coordinate system. Hereinafter, the first point cloud will be referred to as a first ranging point cloud. The first point cloud acquisition unit 15 acquires the first ranging point cloud from the LiDAR device 11P. The first time is an arbitrary time. The first time is preferably a time that can ensure that no foreign object is left behind on the first road surface 5P of the runway 2. However, the first time may be a time that cannot ensure that no foreign object is left behind on the first road surface 5P of the runway 2. FIG. 4 shows a data structure diagram of the first ranging point cloud. As shown in FIG. 4, the first ranging point cloud is composed of a plurality of first ranging points. Each of the first distance measurement points includes a distance measurement point ID and coordinate data of the distance measurement point, the coordinate data being configured of X, Y, and Z values. The first point cloud acquisition unit 15 stores the first distance measurement point cloud in the memory 12b.

[0034] FIG. 5 is a perspective view of the runway 2. As shown in FIG. 5, the runway 2 is inclined in an inverted V shape for drainage purposes. That is, the runway 2 has a first inclined surface 2s and a second inclined surface 2t with a center line 2c as the boundary. The first inclined surface 2s is an inclined surface closer to the LiDAR device 11P than the center line 2c. The second inclined surface 2t is an inclined surface farther from the LiDAR device 11P than the center line 2c. In FIG. 5, a plurality of first ranging points p constituting the first ranging point cloud acquired by the first point cloud acquisition unit 15 are illustrated by small circles. As shown in FIG. 5, the density of the first ranging points p on the first inclined surface 2s is higher than the density of the first ranging points p on the second inclined surface 2t. This is because the angle at which the laser light emitted from the LiDAR device 11P is incident on the second inclined surface 2t is shallower than the angle at which the laser light emitted from the LiDAR device 11P is incident on the first inclined surface 2s.

[0035] The second point cloud acquisition unit 16 acquires a second point cloud generated by measuring the distance to the runway 2 at a second time after the first time. Specifically, the LiDAR device 11P measures the distance to the first road surface 5P of the runway 2 at the second time to generate the second point cloud. The second point cloud is typically a point cloud expressed in an orthogonal coordinate system. Hereinafter, the second point cloud is referred to as a second ranging point cloud. The second point cloud acquisition unit 16 acquires the second ranging point cloud from the LiDAR device 11P. The second time is any time after the first time. The second time is typically a time that occurs at a predetermined interval. The second time is the timing at which the foreign object detection system 10 checks whether or not a foreign object is left behind on the first road surface 5P of the runway 2. FIG. 6 shows a data structure diagram of the second ranging point cloud. As shown in FIG. 6, the second ranging point cloud is composed of a plurality of second ranging points. Each of the second distance measurement points includes a measurement point ID and coordinate data of the corresponding distance measurement point, which is configured of values ​​of X, Y, and Z. The second point cloud acquisition unit 16 stores the second distance measurement point cloud in the memory 12b.

[0036] Fig. 7 is a perspective view of the runway 2. In Fig. 7, similar to Fig. 5, a plurality of second ranging points q constituting the second ranging point cloud acquired by the second point cloud acquisition unit 16 are illustrated by small circles.

[0037] Next, the first calculation unit 17 will be described in detail with reference to Fig. 8 to Fig. 13. The first calculation unit 17 calculates a first point density distribution characteristic in the first detection space based on the first ranging point cloud. Specifically, this is as follows.

[0038] First, the first calculation unit 17 calculates the point density at each of the first ranging points p constituting the first ranging point group. FIG. 8 is a partially enlarged view of FIG. 5. As shown in FIG. 8, the first calculation unit 17 focuses on an arbitrary first ranging point px among the first ranging points p constituting the first ranging point group. Next, the first calculation unit 17 defines a spherical surface s1 having a radius r1 and centered on the first ranging point px. Then, the first calculation unit 17 determines the number of the first ranging points p located inside the spherical surface s1 among the first ranging points p constituting the first ranging point group as the point density at the first ranging point px. In the example of FIG. 8, the point density at the first ranging point px is "9", which is the number of the first ranging points p identified by the thick and small circles. The first calculation unit 17 calculates the above point density for all of the first ranging points p constituting the first ranging point group. Then, the first calculation unit 17 stores the point density PD (point density) calculated for each of the first distance measurement points p constituting the first distance measurement point group in the memory 12b as shown in Fig. 9. As shown in Fig. 9, the point densities PD of the first distance measurement points p constituting the first distance measurement point group are typically different from each other. A larger point density PD means that more first distance measurement points p are concentrated in the vicinity of the first distance measurement point p corresponding to the point density PD.

[0039] The method of calculating the point density by the first calculation unit 17 is not limited to the above calculation method. For example, the first calculation unit 17 defines a sphere centered on the first ranging point px. The first calculation unit 17 gradually increases the radius of the sphere. The first calculation unit 17 may set the reciprocal of the radius of the sphere when the number of the first ranging points p located inside the sphere among the first ranging points p constituting the first ranging point group becomes a predetermined number as the point density at the first ranging point px. Even in this case, a higher point density means that more first ranging points p are concentrated in the vicinity of the first ranging point p corresponding to the point density.

[0040] Next, the first calculation unit 17 generates an inspection point cloud. The inspection point cloud is a point cloud used when the foreign object detection device 12 scans the runway 2 to detect foreign objects. FIG. 10 shows the data structure of the inspection point cloud. As shown in FIG. 10, the first calculation unit 17 typically generates the inspection point cloud in a grid shape such as a ball grid array (BGA) at predetermined intervals in the longitudinal direction of the runway 2 in a planar view and at predetermined intervals in the lateral direction of the runway 2 in a planar view. The Z coordinate of each of the multiple inspection points constituting the inspection point cloud is set to the Z coordinate of the inspection point when the inspection point collides with the surface of the runway 2 when moved in the vertical direction. Here, the surface of the runway 2 can be estimated by RANSAC (Random Sample Consensus) based on the first ranging point cloud.

[0041] Next, the first calculation unit 17 calculates a first point density distribution characteristic at each of the multiple inspection points constituting the inspection point cloud. FIG. 11 is a partially enlarged view of FIG. 5. As shown in FIG. 11, the first calculation unit 17 focuses on an arbitrary inspection point ux among the multiple inspection points constituting the inspection point cloud. Next, the first calculation unit 17 defines a spherical surface s2 having a radius r2 and centered on the inspection point ux. The radius r2 defining the spherical surface s2 is typically larger than the radius r1 defining the spherical surface s1. The detection space sp1 surrounded by the spherical surface s2 corresponds to the above-mentioned first detection space. Next, the first calculation unit 17 calculates a point density distribution characteristic at the inspection point ux based on the point density of the multiple first ranging points p located in the detection space sp1 among the multiple first ranging points p constituting the first ranging point cloud. In FIG. 11, the multiple first ranging points p located in the detection space sp1 are depicted with thick circles. FIG. 12 is a histogram of the point density of the multiple first distance measurement points p located in the detection space sp1. The horizontal axis is the point density, and the vertical axis is the frequency. As shown in FIG. 12, the point density of the multiple first distance measurement points p located in the detection space sp1 is considered to have a generally normal distribution. The point density distribution characteristic at the inspection point ux is composed of the average value (mean) of the point density of the multiple first distance measurement points p located in the detection space sp1 and the variance of the point density of the multiple first distance measurement points p located in the detection space sp1. The average value is a specific example of the first index value. The first index value may be a mode or a median instead of the average value. The first calculation unit 17 calculates the above-mentioned first point density distribution characteristic for all of the multiple inspection points constituting the inspection point cloud. Then, as shown in FIG. 13, the first calculation unit 17 stores the first point density distribution characteristic (m1, σ1) calculated for each of the multiple inspection points constituting the inspection point cloud in the memory 12b. 13, the first point density distribution characteristics of the multiple inspection points constituting the inspection point group are typically different from each other. Since the first point density distribution characteristics are statistical values, they can be said to be an index that is less affected by the density of points caused by the inclination angle of the runway 2.

[0042] Next, the second calculation unit 18 will be described in detail with reference to Fig. 14 to Fig. 18. The second calculation unit 18 calculates a second point density distribution characteristic in a second detection space corresponding to the first detection space, based on the second ranging point cloud.

[0043] First, the second calculation unit 18 calculates the point density at each of the multiple second measurement points q constituting the second measurement point group. FIG. 14 is a partially enlarged view of FIG. 7. As shown in FIG. 14, the second calculation unit 18 focuses on an arbitrary second measurement point qx among the multiple second measurement points q constituting the second measurement point group. Next, the second calculation unit 18 defines a spherical surface s1 having a radius r1 and centered on the second measurement point qx. The radius r1 used when the second calculation unit 18 defines the spherical surface s1 is equal to the radius r1 used when the first calculation unit 17 defines the spherical surface s1. Then, the second calculation unit 18 determines the number of multiple second measurement points q located inside the spherical surface s1 among the multiple second measurement points q constituting the second measurement point group as the point density at the second measurement point qx. In the example of FIG. 14, the point density at the second ranging point qx is "12", which is the number of the second ranging points q identified by the thick small circles. The second calculation unit 18 calculates the point density for all of the second ranging points q constituting the second ranging point group. Then, as shown in FIG. 15, the second calculation unit 18 stores the point density PD calculated for each of the second ranging points q constituting the second ranging point group in the memory 12b. As shown in FIG. 15, the point densities PD of the second ranging points q constituting the second ranging point group are typically different from each other. The larger the point density PD, the more second ranging points q are concentrated in the vicinity of the second ranging point q corresponding to the point density PD.

[0044] The method of calculating the point density by the second calculation unit 18 is not limited to the above calculation method. For example, the second calculation unit 18 defines a spherical surface centered on the second ranging point qx. The second calculation unit 18 gradually expands the radius of the spherical surface. The second calculation unit 18 may set the reciprocal of the radius of the spherical surface when the number of the second ranging points q located inside the spherical surface among the second ranging points q constituting the second ranging point group becomes a predetermined number as the point density at the second ranging point qx. Even in this case, a higher point density means that more second ranging points q are concentrated in the vicinity of the second ranging point q corresponding to the point density.

[0045] Next, the second calculation unit 18 calculates a second point density distribution characteristic at each of the multiple inspection points constituting the inspection point cloud generated by the first calculation unit 17. FIG. 16 is a partially enlarged view of FIG. 7. As shown in FIG. 16, the second calculation unit 18 focuses on an arbitrary inspection point ux among the multiple inspection points constituting the inspection point cloud. Next, the second calculation unit 18 defines a spherical surface s2 having a radius r2 and centered on the inspection point ux. The radius r2 defining the spherical surface s2 is typically larger than the radius r1 defining the spherical surface s1. The detection space sp2 surrounded by the spherical surface s2 corresponds to the above-mentioned second detection space. The detection space sp1 defined by the first calculation unit 17 and the detection space sp2 defined by the second calculation unit 18 are spherical spaces having the same radius. Next, the second calculation unit 18 calculates a point density distribution characteristic at the inspection point ux based on the point density of the multiple second ranging points q located within the detection space sp2 among the multiple second ranging points q constituting the second ranging point cloud. In FIG. 16, the second distance measurement points q located in the detection space sp2 are drawn with thick circles. FIG. 17 is a histogram of the point density of the second distance measurement points q located in the detection space sp2. The horizontal axis is the point density, and the vertical axis is the frequency. As shown in FIG. 17, the point density of the second distance measurement points q located in the detection space sp2 is considered to be approximately normal distribution. The point density distribution characteristic at the inspection point ux is composed of the average value (mean) of the point density of the second distance measurement points q located in the detection space sp2 and the variance of the point density of the second distance measurement points q located in the detection space sp2. The average value is a specific example of the first index value. The first index value may be a mode or a median instead of the average value. The second calculation unit 18 calculates the above-mentioned second point density distribution characteristic for all of the inspection points constituting the inspection point group. Then, the second calculation unit 18 stores the second point density distribution characteristic (m2, σ2) calculated for each of the inspection points constituting the inspection point cloud in the memory 12b, as shown in Fig. 18. As shown in Fig. 18, the second point density distribution characteristics of the inspection points constituting the inspection point cloud are typically different from each other. Since the second point density distribution characteristic is a statistical value, it can be said to be an index that is not easily affected by the density of points caused by the inclination angle of the runway 2.

[0046] In this way, as shown in FIG. 18, the foreign object detection apparatus 12 obtains a first point density distribution characteristic (m1, σ1) corresponding to each of the multiple inspection points that make up the inspection point group, and a second point density distribution characteristic (m2, σ2) corresponding to each of the multiple inspection points that make up the inspection point group.

[0047] Next, the foreign object determination unit 19 determines whether or not a foreign object is left behind on the runway 2 at the second time based on the first point density distribution characteristic and the second point density distribution characteristic. Specifically, the foreign object determination unit 19 determines whether or not a foreign object is left behind on the runway 2 at the second time based on the difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic. More specifically, the foreign object determination unit 19 determines whether or not a foreign object is left behind on the runway 2 at the second time based on a comparison result obtained by comparing the difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic with the second index value of the first point density distribution characteristic. The details are as follows.

[0048] First, the foreign matter determination unit 19 executes the foreign matter determination described above for each of the inspection points constituting the inspection point group.

[0049] For example, the above foreign matter determination is performed on the inspection point with the inspection point ID of 1938. In this case, the foreign matter determination unit 19 compares the average value m1, which is the first index value of the first-point density distribution characteristic, with the average value m2, which is the first index value of the second-point density distribution characteristic. Then, when the difference Δm between the average values ​​m1 and m2 satisfies a predetermined condition, the foreign matter determination unit 19 determines that a foreign matter is left behind at the inspection point with the inspection point ID of 1938. The difference Δm satisfies the predetermined condition typically when the comparison result between the difference Δm and the variance σ1, which is the second index value of the first-point density distribution characteristic, satisfies the predetermined condition. In this embodiment, the difference Δm satisfies the predetermined condition when the following formula (1) is satisfied. However, in the following formula, the coefficient k can be set arbitrarily, typically between 1 and 5. The coefficient k is preferably 3.

[0050]

number

[0051] At the inspection point having an inspection point ID of 1938, Δm is 0.4 and σ1 is 17.2. Therefore, when the coefficient k is set to 3, the above formula (1) does not hold at the inspection point having an inspection point ID of 1938. Therefore, the foreign substance determination unit 19 determines that no foreign substance is left behind at the inspection point having an inspection point ID of 1938 at the second time, which is the distance measurement time of the second distance measurement point group.

[0052] On the other hand, Δm is 52.3 and σ1 is 16.1 at the inspection point whose ID is 1943. Therefore, when the coefficient k is 3, the above formula (1) holds for the inspection point whose ID is 1943. Therefore, the foreign substance determination unit 19 determines that a foreign substance is left behind at the inspection point whose ID is 1943 at the second time, which is the distance measurement time of the second distance measurement point group.

[0053] In short, when the average value m2 deviates significantly from the average value m1, the foreign matter determining unit 19 determines that a foreign matter is left behind at the corresponding inspection point or in its vicinity.

[0054] Alternatively, if the average value m2 deviates significantly from the average value m1, the foreign object determination unit 19 may determine that a foreign object is left behind in the detection space sp1 or the detection space sp2 centered on the corresponding inspection point. If the average value m2 deviates significantly from the average value m1, the foreign object determination unit 19 may determine that a foreign object is left behind at or near the second ranging point q having the highest point density among the multiple second ranging points q in the detection space sp2 centered on the corresponding inspection point. It goes without saying that in this embodiment, the detection space sp1 centered on an arbitrary inspection point and the detection space sp2 centered on that inspection point are the same inspection space.

[0055] Furthermore, the difference Δm satisfying a predetermined condition may be the one satisfying the following formula (2) instead of the above formula (1). However, in the following formula (2), the coefficient g can be set arbitrarily, typically between 1 and 5. The coefficient g is preferably 3.

[0056]

number

[0057] Furthermore, the difference Δm satisfying a predetermined condition may be the one satisfying the following formula (3) instead of the above formula (1). However, in the following formula (3), the coefficient g can be set arbitrarily, typically between 1 and 5. The coefficient g is preferably 3.

[0058]

number

[0059] Next, the output unit 20 outputs an image of the foreign object determination result by the foreign object determination unit 19 to the LCD 12d. An example of the image output is shown in Fig. 19. As shown in Fig. 19, the output unit 20 outputs an image of XYZ coordinates indicating the foreign object detection position to the LCD 12d together with a 3D model of the runway 2 constructed by, for example, RANSAC surface estimation.

[0060] Next, the control flow of the foreign object detector 12 will be described with reference to FIG.

[0061] First, the first point cloud acquisition unit 15 acquires a first ranging point cloud generated by measuring the distance to the runway 2 at a first time (S100, FIGS. 4 and 5).

[0062] Next, the first calculation unit 17 calculates the point density PD at each of the first distance measurement points p that form the first distance measurement point group (S110, FIGS. 8 and 9).

[0063] Next, the first calculation unit 17 generates an inspection point cloud (S120, FIG. 10).

[0064] Next, the first calculation unit 17 calculates a first point density distribution characteristic for each of the plurality of inspection points that make up the inspection point cloud (S130, FIGS. 11, 12, and 13).

[0065] Next, the second point cloud acquisition unit 16 acquires a second ranging point cloud generated by measuring the distance to the runway 2 at a second time (S200, FIGS. 6 and 7).

[0066] Next, the second calculation unit 18 calculates the point density PD at each of the second distance measurement points q that form the second distance measurement point group (S210, FIGS. 14 and 15).

[0067] Next, the second calculation unit 18 calculates the second point density distribution characteristic for each of the plurality of inspection points that make up the inspection point cloud (S220, FIGS. 16, 17, and 18).

[0068] Next, the foreign object determination unit 19 determines, for each of the multiple inspection points that make up the inspection point group, whether or not a foreign object is left behind on the runway 2 at a second time based on the first point density distribution characteristic corresponding to that inspection point and the second point density distribution characteristic corresponding to that inspection point (S300, Figure 18).

[0069] Then, the output unit 20 outputs the result of the foreign matter determination made by the foreign matter determining unit 19 to the LCD 12d as an image (S400, FIG. 19).

[0070] This embodiment has the following features.

[0071] The foreign object detection device 12 is a specific example of a foreign object detection system. The foreign object detection device 12 includes a first point cloud acquisition unit 15, a second point cloud acquisition unit 16, a first calculation unit 17, a second calculation unit 18, and a foreign object determination unit 19. The first point cloud acquisition unit 15 acquires a first ranging point cloud (first point cloud) generated by measuring the distance of the runway 2 (paved road surface) at a first time. The second point cloud acquisition unit 16 acquires a second ranging point cloud (second point cloud) generated by measuring the distance of the runway 2 at a second time. The first calculation unit 17 calculates a first point density distribution characteristic in the detection space sp1 (first detection space) based on the first ranging point cloud. The second calculation unit 18 calculates a second point density distribution characteristic in the detection space sp2 corresponding to the detection space sp1 based on the second ranging point cloud. The foreign object determination unit 19 determines whether or not a foreign object is left behind on the runway 2 at the second time based on the first point density distribution characteristic and the second point density distribution characteristic. The foreign object detection device 12 can detect a foreign object left behind on the runway 2 with a high degree of accuracy.

[0072] That is, on the surface of an object such as the runway 2, the point cloud is generally sampled non-uniformly. Therefore, by focusing on the deviation based on the statistical value of the surrounding point cloud, the foreign object location becomes clearly visible. If the point cloud is sampled uniformly on the surface of the object, the detection accuracy will not change even if the focus is on the point density distribution characteristics instead of the point density. In the case where the laser light is incident at a shallow incidence angle like the runway 2 illustrated in this embodiment and the surface is slightly uneven, or in the case where the scanning method is not an isotropic scanning method such as a conical scan, the density of the point cloud varies depending on the location regardless of the presence or absence of a foreign object. Therefore, it is considered that the detection accuracy of the foreign object is improved by statistically creating a local model and then observing the deviation.

[0073] Moreover, the first point density distribution characteristic and the second point density distribution characteristic each have a first index value indicating either an average value, a mode value, or a median value. The foreign object determination unit 19 determines whether or not a foreign object is left behind on the runway 2 at the second time based on the difference Δm between the first index value m1 of the first point density distribution characteristic and the first index value m2 of the second point density distribution characteristic. The foreign object detection device 12 can detect a foreign object left behind on the runway 2 with even higher accuracy.

[0074] Furthermore, the first point density distribution characteristic has a second index value indicating variance. The foreign object determination unit 19 determines whether or not a foreign object is left behind on the runway 2 at the second time based on a comparison result of a difference Δm between the first index value m1 of the first point density distribution characteristic and the first index value m2 of the second point density distribution characteristic with the second index value σ1 of the first point density distribution characteristic. The foreign object detection device 12 can detect foreign objects left behind on the runway 2 with even higher accuracy.

[0075] The first calculation unit 17 calculates the point density at each of the multiple first ranging points included in the detection space sp1 of the first ranging point group. The first calculation unit 17 calculates a first point density distribution characteristic based on the point density at each of the multiple first ranging points. Similarly, the second calculation unit 18 calculates the point density at each of the multiple second ranging points included in the detection space sp2 of the second ranging point group. The second calculation unit 18 calculates the second point density distribution characteristic based on the point density at each of the multiple second ranging points. The foreign object detection device 12 can calculate the first point density distribution characteristic and the second point density distribution characteristic with a simple calculation.

[0076] Moreover, the detection space sp1 (first detection space) and the detection space sp2 (second detection space) are the same detection space, which allows comparison of point density distribution characteristics in the same detection space.

[0077] Moreover, the detection space sp1 (first detection space) and the detection space sp2 (second detection space) are both spherical spaces, which makes it easier to perform calculations for defining the detection spaces.

[0078] Second embodiment A second embodiment of the present disclosure will be described below with reference to the drawings. The following description will focus on the differences between the second embodiment and the first embodiment, and will not include redundant descriptions.

[0079] In the first embodiment, as shown in Fig. 10, the first calculation unit 17 generated an inspection point cloud. Then, the first calculation unit 17 calculated a first point density distribution characteristic for each of the inspection points constituting the inspection point cloud. Also, the second calculation unit 18 referred to the inspection point cloud generated by the first calculation unit 17. That is, the second calculation unit 18 calculated a second point density distribution characteristic for each of the inspection points constituting the inspection point cloud generated by the first calculation unit 17. Then, the foreign matter determination unit 19 performed foreign matter determination at each of the inspection points constituting the inspection point cloud, based on the first point density distribution characteristic corresponding to the inspection point and the second point density distribution characteristic corresponding to the inspection point.

[0080] In contrast, in this embodiment, the above-mentioned inspection point cloud is not used. Instead, in this embodiment, the first calculation unit 17 calculates a first point density distribution characteristic at each of the multiple first measurement points p that constitute the first measurement point cloud. The calculation method of the first point density distribution characteristic is the same as the calculation method of the first point cloud density distribution characteristic at the inspection point, so a detailed description will be omitted. The first calculation unit 17 stores the first point density distribution characteristic (m1, σ1) calculated for each of the multiple first measurement points p that constitute the first measurement point cloud in the memory 12b, as shown in FIG.

[0081] Similarly, the second calculation unit 18 calculates the second point density distribution characteristic at each of the second measurement points q constituting the second measurement point cloud. The method of calculating the second point density distribution characteristic is the same as the method of calculating the second point cloud density distribution characteristic at the inspection point, so a detailed description will be omitted. The second calculation unit 18 stores the second point density distribution characteristic (m2, σ2) calculated for each of the second measurement points q constituting the second measurement point cloud in the memory 12b, as shown in FIG.

[0082] Next, the foreign object determination unit 19 determines whether or not a foreign object is left behind on the runway 2 at the second time based on the first point density distribution characteristic and the second point density distribution characteristic. Specifically, the foreign object determination unit 19 determines whether or not a foreign object is left behind on the runway 2 at the second time based on the difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic. More specifically, the foreign object determination unit 19 determines whether or not a foreign object is left behind on the runway 2 at the second time based on a comparison result obtained by comparing the difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic with the second index value of the first point density distribution characteristic. The details are as follows.

[0083] First, the foreign object determination unit 19 performs the above foreign object determination for each of the multiple first ranging points p constituting the first ranging point group. In Fig. 21, for example, the foreign object determination is performed for the first ranging point p whose ranging point ID is 283477. In this case, the foreign object determination unit 19 determines whether or not a foreign object is left behind on the runway 2 at the second time based on the first point density distribution characteristic at the first ranging point p whose ranging point ID is 283477 and the second point density distribution characteristic at the second ranging point q that is closest to the first ranging point p in the second ranging point group. Here, for convenience of explanation, it is assumed that the ranging point ID of the second ranging point q that is closest to the first ranging point p whose ranging point ID is 283477 in the second ranging point group shown in Fig. 22 is 319329.

[0084] Therefore, the foreign substance determination unit 19 compares the average value m1, which is the first index value of the first ranging point p whose ranging point ID is 283477, with the average value m2, which is the first index value of the second ranging point q whose ranging point ID is 319329. If the difference Δm between the average values ​​m1 and m2 satisfies a predetermined condition, the foreign substance determination unit 19 determines that a foreign substance is left behind at the first ranging point p whose ranging point ID is 283477. Typically, the difference Δm satisfies the predetermined condition when the comparison result between the difference Δm and the variance σ1, which is the second index value of the first point density distribution characteristic, satisfies the predetermined condition. In this embodiment, the difference Δm satisfies the predetermined condition when the difference Δm satisfies the above-mentioned formula (1). However, instead of this, the difference Δm satisfies the predetermined condition when the difference Δm satisfies the above-mentioned formula (2) or formula (3) may be satisfied.

[0085] In the second ranging point group, the second ranging point q, which is the closest point to the first ranging point p whose ranging point ID is 283477, has a ranging point ID of 319329. Therefore, the difference Δm at the first ranging point p whose ranging point ID is 283477 is -56.0, which is 167.7 minus 221.7, and σ1 is 17.2. Therefore, when the coefficient k is 3, the above formula (1) is established at the first ranging point p whose ranging point ID is 283477. Therefore, the foreign object determination unit 19 determines that a foreign object is left behind at the first ranging point whose ranging point ID is 283477 at the second time, which is the ranging time of the second ranging point group.

[0086] Furthermore, the second ranging point q, which is the closest point to the first ranging point p whose ranging point ID is 283480 in the second ranging point group, has a ranging point ID of 319337. Therefore, the difference Δm at the first ranging point p whose ranging point ID is 283480 is -1.5, which is obtained by subtracting 169.3 from 167.8, and σ1 is 16.2. Therefore, when the coefficient k is 3, the above formula (1) does not hold at the first ranging point p whose ranging point ID is 283480. Therefore, the foreign object determination unit 19 determines that no foreign object is left behind at the first ranging point p whose ranging point ID is 283480 at the second time, which is the ranging time of the second ranging point group.

[0087] In short, if the average value m2 at the second ranging point, which is the closest point to the first ranging point, deviates significantly from the average value m1 at the first ranging point, the foreign object determination unit 19 determines that a foreign object has been left behind at or near the first ranging point.

[0088] Alternatively, the foreign object determination unit 19 may determine that a foreign object is left behind in the detection space sp1 centered on the first distance measurement point when the average value m2 at the second distance measurement point that is the nearest point to the first distance measurement point significantly deviates from the average value m1 at the first distance measurement point. Also, the foreign object determination unit 19 may determine that a foreign object is left behind at or near the second distance measurement point q that has the highest point density among the multiple second distance measurement points q in the detection space sp1 centered on the first distance measurement point when the average value m2 at the second distance measurement point that is the nearest point to the first distance measurement point significantly deviates from the average value m1 at the first distance measurement point.

[0089] Next, the output unit 20 outputs an image of the foreign object determination result by the foreign object determination unit 19 to the LCD 12d. An example of the image output is shown in Fig. 23. As shown in Fig. 23, the output unit 20 outputs an image of XYZ coordinates indicating the foreign object detection position to the LCD 12d together with a 3D model of the runway 2 constructed by, for example, RANSAC surface estimation.

[0090] Next, the control flow of the foreign object detection device 12 will be described with reference to Fig. 24. Below, the control flow shown in Fig. 24 will be described mainly with respect to the parts that differ from the control flow shown in Fig. 20, and overlapping explanations will be omitted as appropriate. The difference between the two is the processing of the thick line blocks in Fig. 24. Moreover, the control flow shown in Fig. 24 does not have step S120 of the control flow shown in Fig. 20.

[0091] In step S130, the first calculation unit 17 calculates the first point density distribution characteristic at each of the multiple first distance measurement points p that form the first distance measurement point group.

[0092] In step S220, the second calculation unit 18 calculates the second point density distribution characteristic at each of the second distance measurement points q that form the second distance measurement point group.

[0093] The foreign substance determination unit 19 performs foreign substance determination at each of the first measurement points p constituting the first measurement point group (S300). In this embodiment, the foreign substance determination unit 19 compares a first point density distribution characteristic in a detection space sp1 centered on an arbitrary first measurement point p with a second point density distribution characteristic in a detection space sp2 centered on a second measurement point q that is the closest point to the first measurement point p among the multiple second measurement points q. Then, the foreign substance determination unit 19 performs foreign substance determination at the first measurement point p based on the comparison result. Therefore, although the detection space sp1 and the detection space sp2 are not the same detection space, they will at least be detection spaces that overlap each other.

[0094] The present disclosure is not limited to the above-described first and second embodiments, and may be modified as appropriate without departing from the spirit and scope of the present disclosure.

[0095] For example, the measurement point group output from the LiDAR device 11 generally includes a reflected luminance value. The reflected luminance value is determined according to the material of the target on which the laser light emitted from the LiDAR device 11 is irradiated. For example, when the laser light emitted from the LiDAR device 11 is irradiated onto the runway 2, the reflected luminance value is relatively low. In contrast, when the laser light emitted from the LiDAR device 11 is irradiated onto metal parts such as bolts and nuts, the reflected luminance value is relatively high. Therefore, when the foreign object determination unit 19 determines a foreign object, the reflected luminance value for each measurement point may be taken into consideration. This will enable foreign object detection to be performed with even higher accuracy when the foreign object is made of metal.

[0096] Furthermore, if the deviation of a measurement point from the runway 2 in the vertical direction is large, there is a high possibility that a foreign object has been left behind on the runway 2 at that measurement point. Therefore, when the foreign object determination unit 19 determines whether or not there is a foreign object, the deviation of each measurement point from the runway 2 in the vertical direction may be taken into consideration. This will achieve a high degree of accuracy in foreign object detection, regardless of the material of the foreign object. The surface of the runway 2 can be estimated using the RANSAC method described above.

[0097] Furthermore, if the degree of accuracy between the normal vector of the measurement point and the normal vector of the runway 2 is high, there is a high possibility that a foreign object is left behind on the runway 2 at the measurement point. Therefore, when the foreign object determination unit 19 determines a foreign object, the angle between the normal vector of the measurement point and the normal vector of the runway 2 may be taken into consideration. This will achieve a high degree of accuracy in foreign object detection regardless of the material of the foreign object. The normal vector of the measurement point may be the normal vector of a surface that can be estimated using RANSAC from multiple measurement points that exist within a predetermined distance from the measurement point. Similarly, the normal vector of the runway 2 may be the normal vector of the runway 2 estimated using the above-mentioned RANSAC.

[0098] In the above example, the program can be stored and provided to the computer using various types of non-transitory computer readable media. The non-transitory computer readable media includes various types of tangible storage media. Examples of the non-transitory computer readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives) and magneto-optical recording media (e.g., magneto-optical disks). Examples of the non-transitory computer readable media further include CD-ROM (Read Only Memory), CD-R, CD-R / W, and semiconductor memory (e.g., mask ROM. Examples of the non-transitory computer readable media further include PROM (Programmable ROM), EPROM (Erasable PROM), flash ROM, and RAM (random access memory). The program may also be provided to the computer by various types of transitory computer readable media. Examples of the transitory computer readable media include electrical signals, optical signals, and electromagnetic waves. The temporary computer-readable medium can supply the program to the computer via a wired communication path such as an electric wire or an optical fiber, or via a wireless communication path.

[0099] A part or all of the above-described embodiments can be described as, but is not limited to, the following supplementary notes.

[0100] (Appendix 1) a first point cloud acquisition means for acquiring a first point cloud generated by measuring the distance to a paved road surface at a first time; a second point cloud acquisition means for acquiring a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time; a first calculation means for calculating a first point density distribution characteristic in a first detection space based on the first point group; a second calculation means for calculating a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point group; a foreign object determination means for determining whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic; Including, Foreign object detection system. (Appendix 2) 2. A foreign object detection system as claimed in claim 1, comprising: each of the first point density distribution characteristic and the second point density distribution characteristic has a first index value indicating any one of an average value, a mode value, or a median value; the foreign object determination means determines whether or not a foreign object is left behind on the paved road surface at the second time based on a difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic. Foreign object detection system. (Appendix 3) 3. A foreign object detection system as recited in claim 2, comprising: the first point density distribution characteristic has a second index value indicative of dispersion; the foreign object determination means determines whether or not a foreign object is left behind on the paved road surface at the second time based on a comparison result of comparing a difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic with the second index value of the first point density distribution characteristic. Foreign object detection system. (Appendix 4) A foreign object detection system according to any one of claims 1 to 3, The first calculation means is Calculating a point density at each of a plurality of points included in the first detection space from the first point group; calculating the first point density distribution characteristic based on the point density at each of the plurality of points; Foreign object detection system. (Appendix 5) A foreign object detection system according to any one of claims 1 to 3, The second calculation means is Calculating a point density at each of a plurality of points included in the second detection space from the second point group; calculating the second point density distribution characteristic based on the point density at each of the plurality of points; Foreign object detection system. (Appendix 6) A foreign object detection system according to any one of claims 1 to 3, The first detection space and the second detection space are the same detection space or detection spaces overlapping each other. Foreign object detection system. (Appendix 7) A foreign object detection system according to any one of claims 1 to 3, The first detection space and the second detection space are spherical spaces. Foreign object detection system. (Appendix 8) a first point cloud acquisition means for acquiring a first point cloud generated by measuring the distance to a paved road surface at a first time; a second point cloud acquisition means for acquiring a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time; a first calculation means for calculating a first point density distribution characteristic in a first detection space based on the first point group; a second calculation means for calculating a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point group; a foreign object determination means for determining whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic; Including, Foreign object detection device. (Appendix 9) 9. A foreign object detection device according to claim 8, each of the first point density distribution characteristic and the second point density distribution characteristic has a first index value indicating any one of an average value, a mode value, or a median value; the foreign object determination means determines whether or not a foreign object is left behind on the paved road surface at the second time based on a difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic. Foreign object detection device. (Appendix 10) 10. A foreign object detection device according to claim 9, the first point density distribution characteristic has a second index value indicative of dispersion; the foreign object determination means determines whether or not a foreign object is left behind on the paved road surface at the second time based on a comparison result of comparing a difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic with the second index value of the first point density distribution characteristic. Foreign object detection device. (Appendix 11) A foreign object detection device according to any one of claims 8 to 10, The first calculation means is Calculating a point density at each of a plurality of points included in the first detection space from the first point group; calculating the first point density distribution characteristic based on the point density at each of the plurality of points; Foreign object detection device. (Appendix 12) A foreign object detection device according to any one of claims 8 to 10, The second calculation means is Calculating a point density at each of a plurality of points included in the second detection space from the second point group; calculating the second point density distribution characteristic based on the point density at each of the plurality of points; Foreign object detection device. (Appendix 13) A foreign object detection device according to any one of claims 8 to 10, The first detection space and the second detection space are the same detection space or detection spaces overlapping each other. Foreign object detection device. (Appendix 14) A foreign object detection device according to any one of claims 8 to 10, The first detection space and the second detection space are spherical spaces. Foreign object detection device. (Appendix 15) The computer A first point cloud is obtained by measuring the distance of a paved road surface at a first time; acquiring a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time; Calculating a first point density distribution characteristic in a first detection space based on the first point group; calculating a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point group; determining whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic; Foreign object detection methods. (Appendix 16) 16. The foreign object detection method according to claim 15, further comprising: each of the first point density distribution characteristic and the second point density distribution characteristic has a first index value indicating any one of an average value, a mode value, or a median value; determining whether or not a foreign object is left behind on the paved road surface at the second time based on a difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic; Foreign object detection methods. (Appendix 17) 17. The foreign object detection method according to claim 16, further comprising: the first point density distribution characteristic has a second index value indicative of dispersion; determining whether or not a foreign object is left behind on the paved road surface at the second time based on a comparison result of comparing a difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic with the second index value of the first point density distribution characteristic; Foreign object detection methods. (Appendix 18) 18. A foreign object detection method according to any one of claims 15 to 17, comprising: Calculating a point density at each of a plurality of points included in the first detection space from the first point group; calculating the first point density distribution characteristic based on the point density at each of the plurality of points; Foreign object detection methods. (Appendix 19) 18. A foreign object detection method according to any one of claims 15 to 17, comprising: Calculating a point density at each of a plurality of points included in the second detection space from the second point group; calculating the second point density distribution characteristic based on the point density at each of the plurality of points; Foreign object detection methods. (Appendix 20) 18. A foreign object detection method according to any one of claims 15 to 17, comprising: The first detection space and the second detection space are the same detection space or detection spaces overlapping each other. Foreign object detection methods. (Appendix 21) 18. A foreign object detection method according to any one of claims 15 to 17, comprising: The first detection space and the second detection space are spherical spaces. Foreign object detection methods. (Appendix 22) The computer A first point cloud is obtained by measuring the distance of a paved road surface at a first time; acquiring a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time; Calculating a first point density distribution characteristic in a first detection space based on the first point group; calculating a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point group; determining whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic; Foreign object detection program. (Appendix 23) 23. A foreign object detection program according to claim 22, each of the first point density distribution characteristic and the second point density distribution characteristic has a first index value indicating any one of an average value, a mode value, or a median value; determining whether or not a foreign object is left behind on the paved road surface at the second time based on a difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic; Foreign object detection program. (Appendix 24) 24. A foreign object detection program according to claim 23, the first point density distribution characteristic has a second index value indicative of dispersion; determining whether or not a foreign object is left behind on the paved road surface at the second time based on a comparison result of comparing a difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic with the second index value of the first point density distribution characteristic; Foreign object detection program. (Appendix 25) 25. A foreign object detection program according to any one of claims 22 to 24, Calculating a point density at each of a plurality of points included in the first detection space from the first point group; calculating the first point density distribution characteristic based on the point density at each of the plurality of points; Foreign object detection program. (Appendix 26) 25. A foreign object detection program according to any one of claims 22 to 24, Calculating a point density at each of a plurality of points included in the second detection space from the second point group; calculating the second point density distribution characteristic based on the point density at each of the plurality of points; Foreign object detection program. (Appendix 27) 25. A foreign object detection program according to any one of claims 22 to 24, The first detection space and the second detection space are the same detection space or detection spaces overlapping each other. Foreign object detection program. (Appendix 28) 25. A foreign object detection program according to any one of claims 22 to 24, The first detection space and the second detection space are spherical spaces. Foreign object detection program. [Explanation of symbols]

[0101] 1 airport 2 Runway 3 Taxiway 4 Tarmac 5P 1st road surface 5Q 2nd road surface 10 Foreign object detection system 11 LiDAR device 11P LiDAR device 11Q LiDAR device 12 Foreign object detection device 13 Approach light 15 1st point cloud acquisition part 16 Second point cloud acquisition section 17 First Calculation Section 18 Second Calculation Section 19 Foreign object determination section 20 Output section p 1st AF point q 2nd AF point sp1 detection space sp2 detection space

Claims

1. a first point cloud acquisition means for acquiring a first point cloud generated by measuring the distance to a paved road surface at a first time; a second point cloud acquisition means for acquiring a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time; a first calculation means for calculating a first point density distribution characteristic in a first detection space based on the first point group; a second calculation means for calculating a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point cloud; a foreign object determination means for determining whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic; Including, Foreign object detection system.

2. 2. The foreign object detection system of claim 1, each of the first point density distribution characteristic and the second point density distribution characteristic has a first index value indicating any one of an average value, a mode value, or a median value; the foreign object determination means determines whether or not a foreign object is left behind on the paved road surface at the second time based on a difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic. Foreign object detection system.

3. 3. The foreign object detection system of claim 2, the first point density distribution characteristic has a second index value indicative of dispersion; the foreign object determination means determines whether or not a foreign object is left behind on the paved road surface at the second time based on a comparison result of comparing a difference between the first index value of the first point density distribution characteristic and the first index value of the second point density distribution characteristic with the second index value of the first point density distribution characteristic. Foreign object detection system.

4. A foreign object detection system according to any one of claims 1 to 3, The first calculation means Calculating a point density at each of a plurality of points included in the first detection space among the first point group; calculating the first point density distribution characteristic based on the point density at each of the plurality of points; Foreign object detection system.

5. A foreign object detection system according to any one of claims 1 to 3, The second calculation means is Calculating a point density at each of a plurality of points included in the second detection space from the second point cloud; calculating the second point density distribution characteristic based on the point density at each of the plurality of points; Foreign object detection system.

6. A foreign object detection system according to any one of claims 1 to 3, The first detection space and the second detection space are the same detection space or detection spaces overlapping each other. Foreign object detection system.

7. A foreign object detection system according to any one of claims 1 to 3, The first detection space and the second detection space are spherical spaces. Foreign object detection system.

8. a first point cloud acquisition means for acquiring a first point cloud generated by measuring the distance to a paved road surface at a first time; a second point cloud acquisition means for acquiring a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time; a first calculation means for calculating a first point density distribution characteristic in a first detection space based on the first point group; a second calculation means for calculating a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point cloud; a foreign object determination means for determining whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic; Including, Foreign object detection device.

9. The computer A first point cloud is obtained by measuring the distance to a paved road surface at a first time; acquiring a second point cloud generated by measuring the distance to the paved road surface at a second time that is later than the first time; Calculating a first point density distribution characteristic in a first detection space based on the first point cloud; calculating a second point density distribution characteristic in a second detection space corresponding to the first detection space based on the second point cloud; determining whether or not a foreign object is left behind on the paved road surface at the second time based on the first point density distribution characteristic and the second point density distribution characteristic; Foreign object detection methods.

10. A foreign object detection program for causing a computer to execute the foreign object detection method according to claim 9.

Images