JP2019046295A - Monitoring device - Google Patents

Abstract

To provide a monitoring device capable of preventing a warning object from being overlooked and preventing an erroneous warning.SOLUTION: A monitoring device includes: a current data calculation part for acquiring distance data till an object present in a monitoring area from a measurement result of a three-dimensional laser scanner for measuring a monitoring area so as to take this data as the current data; a comparison data calculation part for acquiring past distance data from the measurement result so as to convert this data into comparison data; a change area extraction part for calculating a difference value between the current data and the comparison data and extracting the area in which a difference value is equal to or more than a threshold value as a change area candidate; and a 3D mask part for identifying whether a spatial position showing the current data matches a mask position which is the spatial position set on the 3D mask, excluding the change area candidate corresponding to the area of the current data determined as showing the spatial position matching the mask position from the change area candidate, and determining the change area.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a monitoring apparatus that recognizes an object present in a monitoring area.

Conventionally, techniques exist for detecting and recognizing an object using a camera image. Since a camera image is two-dimensional data, information obtained from the camera image is also limited to two-dimensional coordinates, luminance, and color. Most of the prior art uses a luminance background subtraction method to extract a change area in a monitoring area to recognize an object. However, the change area can also be extracted by detecting an object other than the target. A change area extracted by detecting an object or the like other than the object causes an erroneous notification. Therefore, a technique is generally employed to prevent false alarms by shielding unnecessary detection elements with a 2D mask so that objects other than the target objects, which are unnecessary detection elements, are not detected.
For example, Patent Document 1 discloses a monitoring image processing apparatus that uses a comb-shaped mask as a 2D mask to prevent detection of unnecessary detection elements such as a puddle on a road surface. In the device disclosed in Patent Document 1, for example, a small movement such as a water mark of a puddle, which is an unnecessary detection element, and a target object that is a notification target that crosses in front of the puddle (a person in Patent Document 1) "The water mark appears irregularly in a certain place range compared to the person", "the water mark is a relatively small change area compared to the person", and "the water mark Is distinguished by utilizing the difference in the property that it has a shape that is more likely to be separated by the sideways comb-shaped mask, as compared to the person.

Japanese Patent Application Laid-Open No. 10-105687

However, in the related art in which shielding with a 2D mask is performed on a camera image that is two-dimensional data, unnecessary detection elements and a notification target overlap in the field of view of the camera, or both are in the field of view of the camera When it exists in the position which adjoined above, with the unnecessary detection element, even the alerting | reporting object was shielded, and the subject that the alerting | reporting object was overlooked occurred.
Further, even in the technique disclosed in Patent Document 1, when there is no difference in size between the unnecessary detection element and the notification target on the camera image, the difference in properties that can be separated by the comb-shaped mask becomes small, and is unnecessary. There is a problem that the detection element and the notification target can not be distinguished with high accuracy, and the notification target is still overlooked.

  The present invention has been made to solve the problems as described above, and has an object to provide a monitoring device capable of preventing an oversight of a notification target and preventing a false alarm.

  The monitoring apparatus according to the present invention acquires the distance data to the object present in the monitoring area from the measurement result of the three-dimensional laser scanner which measured the monitoring area, and uses the current data operation unit as current data and the measurement result in the past. Distance data is acquired, and a comparison data operation unit that converts the data into comparison data, and a difference value between current data and comparison data are calculated, and a change area extraction that extracts an area having the difference value equal to or more than a threshold as a change area candidate It identifies whether the space position indicated by the current data matches the mask position that is the set space position of the 3D mask, and matches the mask position from the change area candidate extracted by the change area extraction unit A 3D mask unit is provided which excludes the change area candidate corresponding to the area of the current data determined to indicate the spatial position and determines the change area.

  According to the present invention, it is possible to prevent an oversight of a notification target and to prevent an erroneous notification.

FIG. 1 is a block diagram showing a configuration of a monitoring device according to Embodiment 1. It is a figure which shows the structure of a three-dimensional laser scanner. It is explanatory drawing which shows the dispersion | distribution mechanism of a three-dimensional laser scanner. 5 is a flowchart showing the operation of the monitoring device according to the first embodiment. It is a figure for demonstrating the method to avoid a misreport using 2D mask in the image image | photographed with the common camera. In Embodiment 1, it is a figure which shows an example of the three-dimensional model which virtually divided | segmented the visual field of the three-dimensional laser scanner into several grids. In the first embodiment, it is an example of a model showing how a floor, trees and pedestrians in the field of view of a three-dimensional laser scanner look in a grid of 8 × 4 × 4. 7A is a model showing a three-dimensional laser scanner, a positional relationship when the floor, trees and pedestrians are viewed from the side, and FIG. 7B is a model in which 8 × 4 × 4 grids in close contact on the left side in the figure. In the figure, the right side is the model in which the 8 × 4 × 4 grids closely attached shown on the left side of the figure are sliced horizontally into 4 so that all grids can be seen, and FIG. 7C shows 3 It is an image model in which the field of view of a two-dimensional laser scanner is assembled into a virtual image. For example, if a 2D mask is set on a model that shows how floors, trees and pedestrians in the field of view of a three-dimensional laser scanner look in an 8 × 4 × 4 grid, As an example, FIG. 8A is a model showing the positional relationship of a three-dimensional laser scanner, a floor, a tree and a pedestrian as viewed from the side, and FIG. 8B is 8 × 4 × 4 on the left side in the figure. The model in which the grids are in close contact, the model on the right side is the model in which the 8 × 4 × 4 grids are in close contact, shown on the left on the figure, and sliced horizontally into four to make all grids visible FIG. 8C is an image model in which the field of view of the three-dimensional laser scanner is assembled into a virtual image. In the first embodiment, a 3D mask is set to a model showing how a floor, tree and pedestrian in the field of view of a three-dimensional laser scanner look in an 8 × 4 × 4 grid. FIG. 9A is an example of an image, and FIG. 9A is a model showing a positional relationship of a three-dimensional laser scanner, a floor, a tree and a pedestrian viewed from the side, and FIG. 9B is 8 × 4 × 4 on the left side in the figure. A model in which 8 grids are in close contact with each other, and a model in which 8 × 4 × 4 grids are in close contact, shown on the left and on the right in the figure, sliced into 4 in the horizontal direction so that all grids can be seen FIG. 9C is an image model in which the field of view of the three-dimensional laser scanner is assembled into a virtual image. It is a flowchart which shows the determination processing by the recognition process part of the monitoring apparatus which concerns on Embodiment 1. FIG. 11A and 11B are diagrams showing an example of the hardware configuration of the monitoring apparatus according to Embodiment 1 of the present invention.

Hereinafter, in order to explain the present invention in more detail, a mode for carrying out the present invention will be described according to the attached drawings.
Embodiment 1
FIG. 1 is a block diagram showing the configuration of the monitoring apparatus 100 according to the first embodiment.
The monitoring apparatus 100 includes a three-dimensional laser scanner 10, a current data calculation unit 20, a current data storage unit 21, a comparison data calculation unit 30, a comparison data storage unit 31, a change area extraction unit 40, a recognition processing unit 50, and a notification processing unit 60. , And 3D mask unit 70. The 3D mask unit 70 includes a mask identification unit 701 and a rewrite unit 702.
In FIG. 1, a background 200 showing the scanning range of the three-dimensional laser scanner 10 outside the monitoring apparatus 100, an object 201 standing in front of the background 200, and an apparatus above the monitoring apparatus 100 A personal computer (PC) 300 that performs information processing is described. Here, although the apparatus above the monitoring apparatus 100 is the PC 300 as an example, the apparatus above the monitoring apparatus 100 can perform notification processing based on notification processing in the monitoring apparatus 100. For example, it may be an audio output device. The detail of the alerting | reporting process in the monitoring apparatus 100 is mentioned later.

The three-dimensional laser scanner 10 acquires three-dimensional information of an object 201 or the like present in a scan range, and measures the distance to the object 201 or the like.
Here, FIG. 2 is a diagram showing the configuration of the three-dimensional laser scanner 10. As shown in FIG. As shown in FIG. 2, the three-dimensional laser scanner 10 incorporates a laser light emitting unit 11, a dispersion mechanism 13 using a rotating mirror, and a laser light receiving unit 16, scans a range indicated by a background 200 to obtain distance data and intensity Get data The laser light emitting unit 11 emits a laser light pulse 12.
The dispersing mechanism 13 is a mechanism for dispersing the laser light pulse 12 emitted from the laser light emitting unit 11 in the wide angle range. In the example of FIG. 2, the dispersing mechanism 13 using a rotating mirror is shown. Details of the dispersing mechanism 13 using the rotating mirror will be described later. The dispersed laser light pulse 14 dispersed by the dispersing mechanism 13 is irradiated and reflected on the background 200 or an object (not shown in FIG. 2) to form the laser reflected light 15. In the example of FIG. 2, the dispersed laser light pulse 14 is sequentially dispersed and irradiated in the X direction and the Y direction of the background 200. Specifically, it is dispersedly irradiated at six points in the X direction of the background 200, two points in the Y direction of the background 200, and a total of 12 points. Therefore, it can be said that the three-dimensional laser scanner 10 in the example of FIG. 2 has a resolution of 12 pixels in total for 6 pixels in the X direction and 2 pixels in the Y direction.
In addition, although it was set as the dispersion | distribution mechanism 13 which used the rotation mirror in FIG. 2, you may apply the other dispersion | distribution mechanism. For example, a scanless optical system that scans a mirror without a motor may be used.

The laser light receiving unit 16 receives the laser reflected light 15 reflected by the reflection target, calculates the distance to the reflection target based on the time difference from light emission to light reception, and uses it as distance data. In the example of FIG. 2, distances are individually calculated for all the irradiation positions dispersed at 12 points in total, 6 points in the X direction of the background 200 and 2 points in the Y direction of the background 200, and used as distance data. Furthermore, the laser light receiving unit 16 calculates the reflectance at each point to be reflected based on the ratio of the amount of light emitted to the amount of light received for all the dispersed irradiation positions, and uses it as intensity data. The distance data and the intensity data calculated by the laser light receiving unit 16 are output to the current data calculation unit 20 and the comparison data calculation unit 30 shown in FIG.
The distance data and the intensity data for all the irradiation positions calculated by the laser light receiving unit 16 will be referred to as point cloud data 17.
The output of the point cloud data 17 by the laser light receiving unit 16 to the current data operation unit 20 and the comparison data operation unit 30 is performed in frame units. The laser receiving unit 16 is point cloud data 17 obtained by scanning the entire background 200 once, that is, in the example of FIG. 2, a total of 12 points in the X direction of the background 200 and 2 points in the Y direction. The point cloud data 17 obtained by scanning once for each point is output to the current data computing unit 20 and the comparison data computing unit 30 as point cloud data 17 for one frame.

  Next, the details of the dispersing mechanism 13 using a rotating mirror will be described with reference to FIG. The dispersing mechanism 13 includes a first rotating mirror 13a, a first motor 13b, a second rotating mirror 13c, and a second motor 13d. The first rotary mirror 13a operates in synchronization with the pulse frequency of the incident laser light pulse 12, and disperses the laser light pulse 12 in the horizontal direction with respect to the surface of the first rotary mirror 13a. The horizontally dispersed horizontally dispersed laser light pulses 13e are always dispersed at the same angle. The first motor 13 b is a drive source for driving the first rotating mirror 13 a. The second rotating mirror 13c operates in synchronization with the pulse frequency of the incident laser light pulse 12, and further disperses the horizontal dispersion laser light pulse 13e in the vertical direction. Vertically dispersed vertical dispersion laser light pulses 13f are always dispersed at the same angle. The second motor 13d is a drive source for driving the second rotating mirror 13c.

By the above operation, the three-dimensional laser scanner 10 obtains three-dimensional information of X, Y, and Z shown below.
X: Horizontal coordinate Y; Vertical coordinate Z; Distance data Horizontal coordinate X is a value indicating the horizontal position of each pixel, and vertical coordinate Y is a value indicating the vertical position of each pixel. In the example of FIG. 2, the horizontal coordinate X is an integer in the range of 1 to 6, and the vertical coordinate Y is an integer in the range of 1 to 2. The distance data Z is depth information in the Z-axis direction obtained at the position of each pixel. As for the distance data Z, distance data Z measured one by one is obtained for a plurality of pixels specified by all combinations of the horizontal coordinate X and the vertical coordinate Y. In the example of FIG. 2, the corresponding distance data Z is obtained for each of the 12 pixels.
Therefore, with regard to a plurality of combinations of horizontal coordinate X, vertical coordinate Y and distance data Z in three-dimensional information, each combination indicates a specific spatial position.
Since three-dimensional information includes distance data Z, which is depth information in the Z-axis direction, when the object moves in the Z-axis direction on three-dimensional coordinates, that is, toward the three-dimensional laser scanner 10 Even when going straight, the difference can be obtained using the amount of movement in the Z-axis direction.

The current data calculation unit 20 acquires the distance data in the point cloud data 17 output from the three-dimensional laser scanner 10 for all the irradiation positions, and the current distance data in the monitoring area, ie, the measurement range of the three-dimensional laser scanner 10 Are stored in the current data storage unit 21 as current data indicating.
The comparison data calculation unit 30 acquires distance data in the point cloud data 17 output from the three-dimensional laser scanner 10 for all irradiation positions, converts it into comparison data, and stores the data in the comparison data storage unit 31. The conversion processing to comparison data is, for example, processing of obtaining average distance data from the distance data of the past 10 frames retroactively from the acquired distance data and using it as comparison data, or distance data of the frame immediately before the input distance data Can be performed by processing such as obtaining comparison data. These conversion processes are conversion processes for obtaining distance data close to those obtained by scanning only the background 200, excluding the object 201 as much as possible.
The comparison data calculation unit 30 stores distance data acquired from the three-dimensional laser scanner 10 in a data storage unit (not shown), and based on the stored distance data, the distance data traced back in the past You should get it.

The change area extraction unit 40 obtains the current data stored in the current data storage unit 21 and the comparison data stored in the comparison data storage unit 31, and compares the current data with the comparison data in pixel units to obtain a difference value. Is calculated, and a pixel area whose calculated difference value is equal to or greater than a preset threshold value is extracted as a change area candidate. Generally, a fixed threshold value is set, and it is changed and handled as binarized data binarized depending on whether the difference value is the set threshold value or more. In addition, since the current data and the comparison data are configured by distance data, the difference value calculated by the change area extraction unit 40 indicates “a difference in distance”. For example, when the background 200 and the object 201 are included in the current data, and only the background 200 is included in the comparison data, a difference appears only in the pixel corresponding to the object 201, and the obtained difference value is "The distance between the object" is shown. Conversely, when the current data does not include the object 201 and both the current data and the comparison data include only the background 200, the obtained difference value is “0”.
The change area extraction unit 40 extracts a pixel area having a difference value equal to or more than the threshold value as a change area candidate, and outputs information of the change area candidate to the 3D mask unit 70 as change area candidate data. Specifically, for example, the change region candidate data is binarized if each pixel is a change region candidate for a plurality of pixels represented by all combinations of horizontal coordinate X and vertical coordinate Y, for example. If the value of “1” is not a change area candidate, the value of “0” is provided as a binarized difference value.

The 3D mask unit 70 identifies whether the spatial position indicated by the current data matches the mask position. The mask position is a set spatial position of the 3D mask. Then, the 3D mask unit 70 excludes, from the change area candidates extracted by the change area extraction unit 40, the change area candidate corresponding to the area of the current data determined to indicate the spatial position coincident with the mask position. Decide.
Note that “the region of current data determined to indicate a spatial position corresponding to a mask position” is indicated by specifying the position of one or more pixels, and more specifically, one or more Can be specified by a combination of the horizontal coordinate X and the vertical coordinate Y.
The 3D mask is for shielding unnecessary detection elements in the monitoring area, and when the space as the monitoring area is virtually divided into a plurality of cube-shaped grids, for one or more specific grids. Is set in advance. The spatial position corresponding to the grid in which the 3D mask is set is referred to as the set spatial position or mask position of the 3D mask. Details will be described later.

The 3D mask unit 70 includes a mask identification unit 701 and a rewrite unit 702.
The mask identification unit 701 identifies whether the spatial position indicated by the current data matches the mask position. Specifically, the mask identification unit 701 acquires the current data and the setting information of the 3D mask, and the space position indicated by the distance data Z at each coordinate of the current data matches the set space position of the 3D mask. Identify whether or not to Then, the mask identifying unit 701 generates, for all pixels of the current data indicating the mask position, change region zero-value coordinate data including horizontal direction coordinates X and vertical direction coordinates Y indicating the positions of all the pixels.
For example, an operator etc. sets beforehand how to divide the monitoring area into a plurality of grids and to which grid the 3D mask is to be set, and the setting information is illustrated. Not stored in an appropriate storage device. The place where the setting information is stored may be a place where the 3D mask unit 70 can refer to.

  For example, setting information indicating that the monitoring area is divided into eight grids in the X direction, four grids in the Y direction, and four grids in the Z direction is stored in the storage device. In this case, the monitoring area is divided into a total of 128 grids. The 3D mask can be set by, for example, applying a 3D mask status to a specific grid in the 128 grids, and the storage device stores information indicating the specific grid. . In this case, the spatial position corresponding to the grid to which the 3D mask status is given is the mask position.

  The operator or the like can set the 3D mask by an appropriate method. For example, as the simplest specification method, there is a method of directly specifying a grid for which 3D mask setting is desired using 3D mask setting means (not shown). However, since this method designates grids to be set one by one, the work becomes enormous, which may be a burden on the operator or the like.

Therefore, the operator or the like may adopt, for example, a method of setting the 3D mask in two steps of “shape designation” and “specification designation” as a simple method next.
For example, with respect to the 3D mask setting unit, the operator or the like makes the shape of the 3D mask "a rectangular solid" and sets "the position of four bottom grids" and "grids" among a plurality of grids into which the monitoring area is divided. Make an input specifying "height". Further, for example, the operator or the like makes the shape of the 3D mask "sphere" with respect to the 3D mask setting means, and "a position of a grid to be a sphere center point" and "grid Specify the radius as a unit.
The 3D mask setting unit can specify a plurality of grids forming the specified rectangular parallelepiped or sphere based on the input information, and can set the 3D mask. Therefore, according to this method, the operator or the like only needs to perform "shape designation" and "specification designation" as described above, which reduces the work load.
In addition, these are only examples, and if it is a shape whose definition is generally clear such as a triangular pyramid, a cone, a triangular prism, a cylinder, etc., the setting method of the 3D mask by "shape designation" and "specification" is all It is usable. Moreover, it is also possible to assemble arbitrary shapes by these combinations.

The rewriting unit 702 rewrites the difference value of the change area candidate corresponding to the area of the current data, which is determined by the mask identification unit 701 to indicate the space position corresponding to the mask position, to 0. Specifically, the rewrite unit 702 acquires the change area candidate data from the change area extraction unit 40, and acquires the change area zero-value coordinate data from the mask identification unit 701. Then, the rewriting unit 702 rewrites, to “0”, the difference value given to the pixel at the position indicated by the changing area zero value coordinate data among the pixels in the changing area candidate data. The difference value in this case is a binarized difference value. For example, the rewriting unit 702 is a position indicated by the zero-value coordinate data of the change area among the pixels determined as the change area candidate in the change area candidate data. For the pixel of {circle over (1)}, the difference value “1” given to the pixel is rewritten to “0”.
Among the pixels set as the change area candidate in the change area candidate data, a pixel whose difference value is rewritten to “0” by the rewriting unit 702 is excluded from the change area candidates.
The rewriting unit 702 outputs the change area candidate data after rewriting of the difference value to the recognition processing unit 50 as change area data.

  The recognition processing unit 50 extracts feature quantities such as “area”, “horizontal and vertical dimensions”, and “speed” of the change area based on the change area data output by the 3D mask unit 70, and the extracted feature quantities are predetermined. Based on whether the above conditions are satisfied, recognition processing of whether or not the change area is a notification target is performed. The recognition processing unit 50 outputs notification instruction information to the notification processing unit 60 when it recognizes that the change area is a notification target.

  The notification processing unit 60 performs notification processing based on the notification instruction information output from the recognition processing unit 50. Examples of the notification process include a process of transmitting a specific signal to a device such as the higher-order PC 300 or a process of causing the device to sound a buzzer.

  In the first embodiment, as shown in FIG. 1, the three-dimensional laser scanner 10 is provided in the monitoring apparatus 100. However, the present invention is not limited to this. The three-dimensional laser scanner 10 is outside the monitoring apparatus 100. The monitoring device 100 may acquire the point cloud data 17 from the three-dimensional laser scanner 10 via a network or the like.

Next, the operation of the monitoring apparatus 100 according to the first embodiment will be described.
FIG. 4 is a flowchart showing the operation of the monitoring apparatus 100 according to the first embodiment.
Here, in order to simplify the description, the case where the resolution of the three-dimensional laser scanner 10 is 8 × 4 pixels will be described as an example.
First, the three-dimensional laser scanner 10 scans the background 200, that is, the range of the monitoring area (step ST401), and acquires point cloud data 17, that is, distance data and intensity data (step ST402). Specifically, the range of the background 200 is divided into 8 × 4 resolutions of the three-dimensional laser scanner 10 and scanned. The distance data is generally digital data, and here, 8-bit multi-value data per pixel of 8 × 4 pixels.
The three-dimensional laser scanner 10 outputs the acquired point group data to the current data calculation unit 20 and the comparison data calculation unit 30.

The current data computing unit 20 stores the distance data in the point cloud data 17 of 8 × 4 pixels acquired in step ST402 as current data in the current data storage unit 21 (step ST403).
The comparison data calculation unit 30 converts the distance data in the point cloud data 17 of 8 × 4 pixels acquired in step ST 402 in the past and stored in the data storage unit (not shown) into comparison data, and the comparison data storage unit 31. (Step ST404).

  The change area extraction unit 40 obtains the current data stored in the current data storage unit 21 and the comparison data stored in the comparison data storage unit 31, and compares the current data with the comparison data in pixel units to obtain a difference value. Is calculated (step ST405).

The difference value obtained in step ST405 is multi-value data of 8 bits per pixel like the distance data, and the change area extraction unit 40 determines whether the obtained difference value is equal to or more than a preset threshold value. (Step ST406). If the difference value is equal to or larger than the threshold (in the case of “YES” in step ST406), the change area extraction unit 40 extracts the pixel area as a change area candidate (step ST407).
On the other hand, when the difference value is less than the threshold (in the case of “NO” in step ST406), it is determined that the pixel area is not a change area, and the process proceeds to step ST408.

Thereafter, the change area extraction unit 40 determines whether or not the process has been performed for all 8 × 4 pixels (step ST408).
In step ST408, when processing has not been performed for all 8 × 4 pixels (in the case of “NO” in step ST408), the process returns to step ST405, and the above-described process is repeated.
In step ST408, when processing is performed on all 8 × 4 pixels (in the case of “YES” in step ST408), the change area extraction unit 40 uses the information of the extracted change area candidate as change area candidate data as a 3D mask unit. Output to 70.

  The mask identifying unit 701 of the 3D mask unit 70 identifies whether or not each spatial position indicated by the current data matches the set spatial position (mask position) of the 3D mask (step ST409).

  In step ST409, when it is identified that the space position indicated by the current data matches the mask position (in the case of “YES” in step ST409), the rewrite unit 702 of the 3D mask unit 70 selects the mask position among the pixels of the change area candidate. For the pixel corresponding to the pixel of the current data indicating the spatial position that coincides with, the difference value given to the pixel is rewritten to “0” (step ST 410).

  If it is determined in step ST409 that the space position indicated by the current data does not match the mask position (in the case of “NO” in step ST409), the rewrite unit 702 of the 3D mask unit 70 selects the mask of the pixels of the change area candidate. For a pixel corresponding to a pixel of current data indicating a spatial position not coincident with the position, the difference value given to the pixel is used as it is (step ST411).

Thereafter, the mask identifying unit 701 determines whether or not the process has been performed for all 8 × 4 pixels (step ST412).
In step ST412, when processing has not been performed on all 8 × 4 pixels (in the case of “NO” in step ST412), the process returns to step ST409, and the above-described process is repeated.
In step ST412, when processing has been performed for all 8 × 4 pixels (in the case of “YES” in step ST412), rewriting section 702 uses the change area candidate data after rewriting the difference value as the change area data. , To the recognition processing unit 50.

Here, the effect of the process in which the 3D mask unit 70 rewrites the difference value in steps ST409 to ST412 (hereinafter, referred to as “3D mask process”) will be described with reference to the drawings.
First, FIG. 5 is a diagram for describing a method of avoiding false alarm using a 2D mask in an image captured by a general camera. FIG. 5A shows, for example, an image of the field of view of a general camera when a floor 501 and a tree 502 are taken as the background 200, and a pedestrian 503 is taken as the moving object 201. In the example, it is assumed that the tree 502 is swaying in the wind and is misidentified as a rare moving object, causing a false alarm. FIG. 5B is a diagram showing an example of an image in the case where a 2D mask 504 is set in order to avoid a false notification due to the tree 502 in the image obtained by capturing the visual field as shown in FIG. 5A.
The 2D mask is a mask that blocks the field of view on the camera view, and the portion hidden by the 2D mask is electrically shielded and can not be seen.
In FIG. 5B, a 2D mask 504 is set to hide the tree 502 that is the cause of the false alarm. As a result, the tree 502 can not be seen on the image and will not cause false alarms.

  However, as shown in FIG. 5B, the 2D mask 504 partially shields the pedestrian 503 simultaneously with the tree 502. Therefore, when the pedestrian 503 moves the position of the 2D mask 504 on the image, it is difficult to find the pedestrian 503, and an oversight of a notification target (hereinafter referred to as “missing notification”) may occur.

Therefore, in the monitoring device 100 according to the first embodiment, the 3D mask unit 70 performs 3D mask processing in the field of view of the 3D laser scanner to distinguish and shield the tree 502 and the object 201 from each other. In addition to preventing a false alarm on the object 201, a false alarm on the tree 502 is prevented.
The details of the 3D mask processing will be described below.

FIG. 6 is a diagram showing an example of a three-dimensional model in which the field of view of the three-dimensional laser scanner 10 is virtually divided into a plurality of grids in the first embodiment.
In FIG. 6, one cube-shaped grid indicates one measurement point of laser irradiation, and a space which is a monitoring area is a total of 128 X grids x Y grids 4 grids x Z grids 4 grids. It is assumed that it is divided into grids.

FIG. 7 shows how the floor 501, the tree 502 and the pedestrian 503 in the field of view of the three-dimensional laser scanner 10 look in the 8 × 4 × 4 grid in the first embodiment. It is an example of a model.
FIG. 7A is a model showing the positional relationship of the three-dimensional laser scanner 10, the floor 501, the tree 502 and the pedestrian 503 as viewed from the side. In FIG. 7B, the left side shows a model in which 8 × 4 × 4 grids are in close contact, and the right side shows a model in which 8 × 4 × 4 grids are close to each other. It is a model that is sliced to four horizontally to make the whole grid visible. FIG. 7C is an image model in which the field of view of the three-dimensional laser scanner 10 is assembled into a virtual image.

  As shown in FIG. 7C, the floor 501 appears to be at a higher position as it goes further by the depression angle of the three-dimensional laser scanner 10. Also, although the tree 502 and the pedestrian 503 are the same size, the tree 502 in the back is measured small.

  7A to 7B, the grids other than the grids corresponding to the floor 501, the tree 502, and the pedestrian 503 are grids for which the three-dimensional laser scanner 10 can not obtain a distance. In the first embodiment, a grid whose distance can not be obtained in the three-dimensional laser scanner 10 is also referred to as a blank grid 505.

As shown in FIGS. 6 and 7, in the field of view of the three-dimensional laser scanner 10, a tree 502 constituting a part of the background 200 overlaps a pedestrian 503 which is the object 201, but the moving object It is necessary to set the pedestrian 503 that is the target as a notification target and exclude the tree swaying in the wind from the target.
However, the information obtained from the three-dimensional laser scanner 10 alone does not include information as to whether the distance data Z at each pixel of the point cloud data 17 corresponds to the pedestrian 503 or the tree 502, and thus the walking Both the person 503 and the tree 502 can be notified. On the other hand, assuming that the 2D mask 504 as described with reference to FIG. 5 is set in order to shield the tree 502 that does not need to be notified, each grid on the field of view of the three-dimensional laser scanner 10 is shown in FIG. It will be.

FIG. 8 shows, for example, a 2D model showing how a floor 501, trees 502 and pedestrians 503 in the field of view of the three-dimensional laser scanner 10 look like in an 8 × 4 × 4 grid. It is an example of the image at the time of setting the mask 504. FIG.
FIG. 8A is a model showing the positional relationship of the three-dimensional laser scanner 10, the floor 501, the tree 502 and the pedestrian 503 as viewed from the side. In FIG. 8B, the left side shows a model in which 8 × 4 × 4 grids are in close contact, and the right side shows a model in which 8 × 4 × 4 grids are in close contact. It is a model that is sliced to four horizontally to make the whole grid visible. Further, FIG. 8C is an image model in which the field of view of the three-dimensional laser scanner 10 is assembled into a virtual image.

In FIG. 8B, it is assumed that the 2D mask 504 is virtually expanded in an 8 × 4 × 4 grid. The 2D mask 504 is defined in the plane of the XY direction, and has no information in the Z direction. Therefore, as shown in FIG. 8B, the 2D mask 504 has a field of view at all positions in the Z direction in the region of the XY direction in which the 2D mask 504 is set, from the near side to the far side as viewed from the three-dimensional laser scanner 10. To block
In this case, as shown in FIG. 8C, when the pedestrian 503 moves the position of the 2D mask 504, a part of the pedestrian 503 is also shielded, which may cause a false alarm.

  Therefore, in the first embodiment, as described above, the 3D mask unit 70 performs 3D mask processing to distinguish and shield the background 200 and the object 201 from each other.

FIG. 9 shows how the floor 501, the tree 502 and the pedestrian 503 in the field of view of the three-dimensional laser scanner 10 look in the 8 × 4 × 4 grid in the first embodiment. It is an example of the image which set 3D mask 506 to the model.
FIG. 9A is a model showing the positional relationship of the three-dimensional laser scanner 10, the floor 501, the tree 502 and the pedestrian 503 as viewed from the side. Further, FIG. 9B shows a model in which 8 × 4 × 4 grids are in close contact with the left side in the figure, and a model in which 8 × 4 × 4 grids are in close contact with the right side in the figure. It is a model that is sliced to four horizontally to make the whole grid visible. Further, FIG. 9C is an image model in which the field of view of the three-dimensional laser scanner 10 is assembled into a virtual image.

FIG. 9B shows an example of an image in which the 3D mask 506 is set in an 8 × 4 × 4 grid. Specifically, in FIG. 9B, it is assumed that the 3D mask 506 is previously set at a position of the tree 502 in the field of view of the three-dimensional laser scanner 10 as a rectangular solid consisting of 12 grids so as to shield the tree 502. .
Unlike the 2D mask 504, the 3D mask 506 can be set to only the grid intended by the user of the monitoring apparatus among a plurality of grids obtained by dividing the space of the monitoring area in the X, Y, and Z directions. In the field of view of the three-dimensional laser scanner 10, it is possible to block only the field of view at a part of the Z direction position intended by the user.

  In FIG. 9B, 3D mask processing is performed only at the position where the tree 502 exists in the Z direction of the field of view of the three-dimensional laser scanner 10. Therefore, although the tree 502 is shielded, the pedestrian 503 present on the near side in the Z direction with respect to the set spatial position of the 3D mask is not shielded (see also FIG. 9C).

  As described above, in the first embodiment, the background 200 and the target object 201 can be distinguished and shielded by the 3D mask 506 by the 3D mask unit 70 performing the 3D mask processing. That is, for example, even if the pedestrian 503 which is the target object 201 to be notified and the tree 502 to be out of the target object are overlapped in the field of view of the three-dimensional laser scanner 10 Deviation can be used to shield only the tree 502 that causes false alarms with the 3D mask 506. This makes it possible to accurately identify the pedestrian 503 to be notified and prevent a false alarm, and to prevent false notification due to the tree 502 to be excluded from being a false alarm, as a false alarm. can do.

It returns to the flowchart of FIG.
When processing is performed on all 8 × 4 pixels in step ST412 (in the case of “YES” in step ST412), the recognition processing unit 50 changes based on the change area data output by the 3D mask unit 70 in step ST412. It is determined whether the area satisfies the matching condition (step ST413).
If the matching condition is satisfied (in the case of “YES” in step ST413), the recognition processing unit 50 recognizes that the change area is a notification target (step ST414). The recognition processing unit 50 outputs notification instruction information to the notification processing unit 60 when it recognizes that the change area is a notification target.
On the other hand, when the matching condition is not satisfied (in the case of “NO” in step ST413), it is determined that the change area is not a notification target (step ST415), and the process returns to step ST401.

Here, the determination processing by the recognition processing unit 50 in step ST413 will be described in detail.
FIG. 10 is a flowchart showing determination processing by the recognition processing unit 50 of the monitoring apparatus 100 according to the first embodiment.
The recognition processing unit 50 determines whether or not the change area exists in the monitoring range (step ST1001). Here, the monitoring range refers to a range within the monitoring area, for example, a range in which notification is required when the object 201 is detected for monitoring, and the monitoring range is set in advance. It shall be done.
When the change area exists in the monitoring range (in the case of “YES” in step ST1001), the recognition processing unit 50 further determines whether the change area has a predetermined area (step ST1002). The term "change region" as used herein refers to a region consisting of a set of a plurality of pixels adjacent to or adjacent to each other among the pixels determined as the change region in step ST411 in FIG.

  When the change area has a predetermined area (in the case of “YES” in step ST 1002), the recognition processing unit 50 further determines whether the change area has a predetermined vertical and horizontal dimension (step ST 1003). If it has predetermined vertical and horizontal dimensions (in the case of "YES" in step ST1003), the recognition processing unit 50 further determines whether or not the change area has a predetermined moving speed (step ST1004). . If it has a predetermined moving speed (in the case of “YES” in step ST1004), the process proceeds to step ST414 in FIG. 4 and it is recognized that the change area is a notification target.

  On the other hand, if the change area does not exist in the monitoring range ("NO" in step ST1001), or if the change area does not have a predetermined area ("NO" in step ST1002), or When the change area does not have predetermined vertical and horizontal dimensions (in the case of "NO" in step ST1003) or in the case where the change area does not have a predetermined moving speed (in the case of "NO" in step ST1004) The process proceeds to step ST415 in FIG. 4 and is determined not to be a notification target.

It returns to the flowchart of FIG.
The notification processing unit 60 performs notification processing for the recognized notification target based on the notification instruction information output from the recognition processing unit 50 in step ST414 (step ST416), and returns to the process of step ST401.

11A and 11B are diagrams showing an example of the hardware configuration of the monitoring apparatus 100 according to the first embodiment.
In the first embodiment of the present invention, the functions of the current data calculation unit 20, the comparison data calculation unit 30, the change area extraction unit 40, the recognition processing unit 50, the notification processing unit 60, and the 3D mask unit 70 , And implemented by the processing circuit 1101. That is, the monitoring apparatus 100 includes a processing circuit 1101 for performing notification processing when a change to be notified is detected based on point cloud data acquired from the three-dimensional laser scanner 10.
The processing circuit 1101 may be dedicated hardware as shown in FIG. 11A or a CPU (Central Processing Unit) 1106 that executes a program stored in the memory 1105 as shown in FIG. 11B.

  When the processing circuit 1101 is dedicated hardware, the processing circuit 1101 may be, for example, a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, an application specific integrated circuit (ASIC), an FPGA (field-programmable) Gate Array) or a combination thereof is applicable.

  When the processing circuit 1101 is the CPU 1106, each function of the current data operation unit 20, the comparison data operation unit 30, the change area extraction unit 40, the recognition processing unit 50, the notification processing unit 60, and the 3D mask unit 70 is It is realized by software, firmware, or a combination of software and firmware. That is, the current data operation unit 20, the comparison data operation unit 30, the change area extraction unit 40, the recognition processing unit 50, the notification processing unit 60, and the 3D mask unit 70 are HDD (Hard Disk Drive) 1102, memory This is realized by a CPU 1106 that executes a program stored in 1105 or the like, and a processing circuit such as a system LSI (Large-Scale Integration). The programs stored in the HDD 1102, the memory 1105, etc. are the current data calculation unit 20, the comparison data calculation unit 30, the change area extraction unit 40, the recognition processing unit 50, the notification processing unit 60, and the 3D mask unit. It can also be said that the computer executes the 70 procedures and methods. Here, the memory 1105 is, for example, a nonvolatile memory such as a random access memory (RAM), a read only memory (ROM), a flash memory, an erasable programmable read only memory (EPROM), and an electrically erasable programmable read only memory (EEPROM). Semiconductor memory, magnetic disk, flexible disk, optical disk, compact disk, mini disk, DVD (Digital Versatile Disc), and the like.

In addition, the hardware dedicated to a part of each function of the current data calculation unit 20, the comparison data calculation unit 30, the change area extraction unit 40, the recognition processing unit 50, the notification processing unit 60, and the 3D mask unit 70. It may be realized by hardware, and a part may be realized by software or firmware. For example, the functions of the current data calculation unit 20 and the comparison data calculation unit 30 are realized by the processing circuit 1101 as dedicated hardware, and the change area extraction unit 40, the recognition processing unit 50, and the notification processing unit 60. The function of the 3D mask unit 70 can be realized by the processing circuit reading and executing a program stored in the memory 1105.
The current data storage unit 21 and the comparison data storage unit 31 use, for example, the HDD 1102. This is merely an example, and the current data storage unit 21 and the comparison data storage unit 31 may be configured by a DVD, a memory 1105, and the like.
The monitoring device 100 further includes an input interface device 1103 and an output interface device 1104 that communicate with the three-dimensional laser scanner 10 or a device such as the PC 300 located above.

  As described above, according to the first embodiment, the monitoring apparatus 100 acquires the distance data to the object present in the monitoring area from the measurement result of the three-dimensional laser scanner 10 for which the monitoring area has been measured. The difference data between the current data and the comparison data, and calculating the difference value between the current data and the comparison data, and the difference value is the threshold value. A change area extraction unit 40 that extracts the above area as a change area candidate, and whether or not the spatial position indicated by the current data matches the mask position that is the set spatial position of the 3D mask is identified. 3D mask which excludes the change area candidate corresponding to the area of the present data determined to indicate the spatial position coincident with the mask position from the change area candidates extracted by the extraction unit 40, and determines the change area Because you configured with 70, oversight of the origination report object caused by the difference in perspective between the unwanted sensing elements and alarm object, and can prevent false alarms.

In the first embodiment described above, the monitoring apparatus 100 includes the three-dimensional laser scanner 10. However, the monitoring apparatus 100 is not limited to the three-dimensional laser scanner 10, and can be applied to general sensors that measure distance.
In the first embodiment described above, the change area extraction unit 40 performs the binarization process and handles the difference value as binarized data. However, the binarization process of the difference value is performed by the change area extraction unit. Instead of being performed by 40, instead, the recognition processing unit 50 may be performed, or the 3D mask unit 70 may be performed. When the calculation amount does not cause a problem as a load, the monitoring apparatus 100 may not perform the binarization process, and may perform the process in each part using the difference value as it is.

  Further, within the scope of the invention of the present application, it is possible to deform any component of the embodiment or omit any component of the embodiment.

  10 three-dimensional laser scanner 11 laser light emitting unit 12 laser light pulse 13 dispersion mechanism 14 dispersion laser light pulse 15 laser reflected light 16 laser light receiving unit 17 point group data 20 actual data operation unit 21 actual data Storage unit, 30 comparison data calculation unit, 31 comparison data storage unit, 40 change area extraction unit, 50 recognition processing unit, 60 notification processing unit, 70 3D mask unit, 100 monitoring device, 200 background, 201 object, 300 PC, 501 floor 502 tree 503 pedestrian 504 2D mask 505 blank grid 506 3D mask 701 mask identification unit 702 rewriting unit 1101 processing circuit 1102 HDD 1103 input interface device 1104 output interface device 1105 memory , 110 CPU.

Claims (3)

A current data calculation unit which acquires distance data to an object present in the monitoring area from the measurement results of the three-dimensional laser scanner which measured the monitoring area, and uses it as current data;
A comparison data operation unit that acquires the past distance data from the measurement result and converts it into comparison data;
A change area extraction unit which calculates a difference value between the current data and the comparison data and extracts an area having the difference value equal to or more than a threshold as a change area candidate;
It identifies whether or not the spatial position indicated by the current data matches the mask position which is the set spatial position of the 3D mask, and matches the mask position from the change area candidate extracted by the change area extraction unit A monitoring apparatus comprising: a 3D mask unit which excludes a change area candidate corresponding to the area of the current data determined to indicate a spatial position and determines a change area. The 3D mask unit is
A mask identification unit that identifies whether or not each spatial position indicated by the current data matches the mask position;
A rewriting unit that rewrites the difference value of the change area candidate corresponding to the area of the current data identified as indicating a space position coincident with the mask position by the mask identification unit to 0; The monitoring device according to 1. The recognition process part which performs recognition processing of whether the said change area is alerting | reporting object based on the feature-value of the change area which the said 3D mask part determined was provided. It was characterized by the above-mentioned. Monitoring equipment.

Images