JP6309099B2 - Monitoring device - Google Patents

Description

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

  Conventionally, there is a technique 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, color, and the like. Many of the prior arts recognize a target object by extracting a change area in a monitoring area using a luminance background difference method. For example, in the monitoring image processing apparatus disclosed in Patent Document 1, a change area is extracted by performing a comparison operation on the current image data and the comparison image data, and when the extracted change area meets a predetermined condition, When the change area appears in a predetermined area when it is recognized as a report target, the process of the report target is restricted when it is determined that a disturbance has occurred.

Japanese Patent Laid-Open No. 9-214939

  However, in the technique described in Patent Document 1 described above, there is a problem that accuracy for monitoring a three-dimensional monitoring region is inferior in two-dimensional data obtained from a camera image. In addition, when there is a luminance difference between the object and the background, a luminance background difference is effectively obtained, but when there is no luminance difference between the object and the background, the object can be accurately recognized. There was a problem that it was not possible. Furthermore, there is a sufficient luminance difference between the object and the background, but when the object moves in the Z-axis direction on the three-dimensional coordinate (goes straight toward the imaging means such as a camera), the two-dimensional coordinate It is recognized that the object is stationary, and it is difficult to recognize the object as a report target. As described above, there is a problem that even if there is a sufficient luminance difference between the object and the background, the object cannot be accurately recognized.

  The present invention has been made to solve the above-described problems. When there is no sufficient luminance difference between the object and the background, or when the object moves on the three-dimensional coordinate in the Z-axis direction. The object of the present invention is to provide a monitoring device capable of accurately recognizing an object.

The monitoring device according to the present invention acquires the distance information to the monitoring area from the measurement result of the three-dimensional laser scanner that measured the monitoring area, and sets the current data calculation unit as the current distance data, and the measurement result to the monitoring area. The difference value between the comparison data calculation unit that acquires distance information and converts it into comparison distance data, the current distance data acquired by the current data calculation unit, and the comparison distance data converted by the comparison data calculation unit is calculated, and the difference A first change region extraction unit that extracts a region having a value equal to or greater than a threshold value as a change region, and the first change region extraction unit should reject based on a distance relationship between the background and the object in the monitoring region and map data stored difference value, and compares the difference value between the comparison distance data and current distance data is shall comprise positive and negative map matching unit for correcting the difference value being rejected by the map data.

  According to the present invention, even when there is no sufficient luminance difference between the object and the background, or even when the object moves on the three-dimensional coordinates, the object can be accurately recognized.

1 is a block diagram illustrating a configuration of a monitoring device according to Embodiment 1. FIG. It is a figure which shows the structure of a three-dimensional laser scanner. It is explanatory drawing which shows the dispersion mechanism of a three-dimensional laser scanner. 3 is a flowchart showing an operation of the monitoring apparatus according to the first embodiment. 4 is a flowchart illustrating a determination process of a recognition processing unit of the monitoring apparatus according to the first embodiment. FIG. 6 is a block diagram illustrating a configuration of a monitoring device according to a second embodiment. 10 is a flowchart showing the operation of the monitoring apparatus according to the second embodiment. FIG. 10 is a block diagram illustrating a configuration of a monitoring device according to a third embodiment. FIG. 10 is a diagram illustrating an example of processing of a distance calculation unit of a monitoring device according to Embodiment 3. It is a figure which shows extraction of the change area | region of the monitoring apparatus which concerns on Embodiment 3. FIG. 10 is a flowchart showing the operation of the monitoring apparatus according to the third embodiment. 10 is a flowchart illustrating a determination process of a recognition processing unit of the monitoring apparatus according to the third embodiment. FIG. 10 is a block diagram illustrating a configuration of a monitoring device according to a fourth embodiment. 10 is a flowchart showing the operation of the monitoring apparatus according to the fourth embodiment. FIG. 10 is a block diagram illustrating a configuration of a monitoring device according to a fifth embodiment. It is a figure which shows an example of a process of the intensity | strength calculating part of the monitoring apparatus which concerns on Embodiment 5. FIG. It is a figure which shows extraction of the change area | region of the monitoring apparatus which concerns on Embodiment 5. FIG. 10 is a flowchart showing the operation of the monitoring apparatus according to the fifth embodiment. FIG. 10 is a block diagram illustrating a configuration of a monitoring device according to a sixth embodiment. It is a figure which shows the process of the stability confirmation part of the monitoring apparatus which concerns on Embodiment 6. FIG. 14 is a flowchart illustrating the operation of the monitoring apparatus according to the sixth embodiment. 18 is a flowchart illustrating a determination process of a recognition processing unit of the monitoring apparatus according to the sixth embodiment. It is a block diagram which shows the structure of the monitoring apparatus implement | achieved combining the structure of Embodiment 1 and Embodiment 3. FIG. It is a block diagram which shows the structure of the monitoring apparatus implement | achieved combining the structure of Embodiment 1 and Embodiment 5. FIG. It is a block diagram which shows the structure of the monitoring apparatus implement | achieved combining the structure of Embodiment 3 and Embodiment 5. FIG. It is a block diagram which shows the structure of the monitoring apparatus implement | achieved combining the structure of Embodiment 1, Embodiment 3, and Embodiment 5. FIG. FIG. 10 is a block diagram illustrating a configuration of a monitoring device according to a seventh embodiment. 18 is a flowchart showing the operation of the monitoring apparatus according to the seventh embodiment. FIG. 20 is a block diagram illustrating a configuration of a monitoring device according to an eighth embodiment. 20 is a flowchart showing the operation of the monitoring apparatus according to the eighth embodiment.

Hereinafter, in order to explain the present invention in more detail, modes for carrying out the present invention will be described with reference to the accompanying drawings.
Embodiment 1 FIG.
FIG. 1 is a block diagram illustrating a configuration of the monitoring apparatus 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 first change area extraction unit 40, a recognition processing unit 50, and a notification. The processing unit 60 is configured. In FIG. 1, a background 200 indicating a scanning range of the three-dimensional laser scanner 10 and an object 201 standing in front of the background 200 are shown outside the monitoring apparatus 100.

FIG. 2 is a diagram showing the configuration of the three-dimensional laser scanner. As shown in FIG. 2, the three-dimensional laser scanner 10 includes a laser light emitting unit 11, a dispersion mechanism 13 using a rotating mirror, and a laser light receiving unit 16, and scans the range indicated by the background 200 to obtain distance data and intensity. Get the data. The laser light emitting unit 11 irradiates a laser light pulse 12.
The dispersion mechanism 13 is a mechanism that disperses the laser light pulse 12 emitted from the laser light emitting unit 11 in a wide angle range. In the example of FIG. 2, a dispersion mechanism 13 using a rotating mirror is shown. Details of the dispersion mechanism 13 using the rotating mirror will be described later. The dispersed laser light pulse 14 dispersed by the dispersion mechanism 13 is irradiated and reflected on the background 200 or an object (not shown in FIG. 2) to form a laser reflected light 15. In the example of FIG. 2, the dispersion laser light pulse 14 is sequentially dispersedly irradiated in the X direction and the Y direction of the background 200. Specifically, 6 points in the X direction of the background 200 and 2 points in the Y direction of the background 200 are distributedly irradiated to a total of 12 points.

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 sets it as distance data. In the example of FIG. 2, the distance is calculated individually for all irradiation positions distributed in a total of 12 points, 6 points in the X direction of the background 200 and 2 points in the Y direction of the background 200. Further, the laser light receiving unit 16 calculates the reflectance at each point of the reflection target based on the ratio of the irradiated light amount and the received light amount with respect to all the dispersed irradiation positions, and sets it as intensity data. The distance data and intensity data 17 calculated by the laser receiving unit 16 are output to the current data calculation unit 20 and the comparison data calculation unit 30 shown in FIG.
In FIG. 2, the dispersion mechanism 13 using a rotating mirror is used, but other dispersion mechanisms may be applied. For example, a scanless optical system that scans a mirror without a motor may be used.

  Next, details of the dispersion mechanism 13 using a rotating mirror will be described with reference to FIG. The dispersion mechanism 13 includes a first rotating mirror 13a, a first motor 13b, a second rotating mirror 13c, and a second motor 13d. The first rotating 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 rotating mirror 13a. Horizontally dispersed laser light pulses 13e dispersed in the horizontal direction are always dispersed at the same angle. The first motor 13b is a drive source that drives the first rotating mirror 13a. 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. The vertically dispersed laser light pulses 13f dispersed in the vertical direction are always dispersed at the same angle. The second motor 13d is a drive source that drives the second rotating mirror 13c.

With the above operation, the three-dimensional laser scanner 10 obtains the following three-dimensional information of X, Y, and Z.
X: horizontal coordinate (6 points in the example of FIG. 2)
Y: vertical coordinate (2 points in the example of FIG. 2)
Z: Distance data (depth information in the Z-axis direction (hereinafter referred to as Z-axis information))
Since the three-dimensional information includes the Z-axis information, the amount of movement in the Z-axis direction can be determined even when the object moves in the Z-axis on the three-dimensional coordinates (goes straight toward the three-dimensional laser scanner 10). To obtain the difference.

  The current data calculation unit 20 acquires distance data input from the three-dimensional laser scanner 10 and stores the current distance data of the monitoring area in the current data storage unit 21 as current data. The current data calculation unit 20 often stores the input distance data itself in the current data storage unit 21. The comparison data calculation unit 30 acquires distance data input from the three-dimensional laser scanner 10, converts it into comparison data, and stores it in the comparison data storage unit 31. For example, the conversion processing to the comparison data is performed by obtaining the average distance data from the distance data for the past 50 frames from the input distance data and using it as comparison data, or the distance data of the frame immediately before the input distance data. Obtained as comparison data.

  The first change area extraction unit 40 acquires 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 and the comparison data in units of pixels. Then, a difference value is calculated, and a pixel area whose calculated difference value is equal to or larger than a preset threshold value is extracted as a change area. Generally, a fixed threshold value is set, and converted into binary data depending on whether or not the difference value is greater than or equal to the set threshold value.

  The recognition processing unit 50 recognizes whether or not the change area is a notification target based on whether or not the condition of the change area extracted by the first change area extraction unit 40 satisfies a predetermined condition. Do. The notification processing unit 60 performs notification processing based on the recognition result of the recognition processing unit 50. Examples of the notification process include a process of transmitting a specific signal to a higher-level PC or the like, or a process of sounding a buzzer of the apparatus.

Next, the operation of the monitoring device 100 will be described.
FIG. 4 is a flowchart showing the operation of the monitoring apparatus according to the first embodiment.
In the flowchart of FIG. 4, the case where the resolution of the three-dimensional laser scanner 10 is 80 × 60 pixels will be described as an example.
First, the three-dimensional laser scanner 10 scans the range of the background 200 (step ST1), and acquires distance data and intensity data (step ST2). Specifically, the range of the background 200 is divided into 80 × 60, which is the resolution of the three-dimensional laser scanner 10, and scanned. The distance data is generally digital data, and here, multi-value data of 8 bits per pixel of 80 × 60 pixels is used. The current data calculation unit 20 accumulates the 80 × 60 pixel distance data acquired in step ST2 in the current data accumulation unit 21 as current data (step ST3). The comparison data calculation unit 30 converts the distance data of 80 × 60 pixels acquired in step T2 into comparison data and stores it in the comparison data storage unit 31 (step ST4).

  The first change area extraction unit 40 calculates a difference value for each pixel by using the current data stored in the current data storage unit 21 and the comparison data stored in the comparison data storage unit 31 (step ST5). ). In the process of step ST5, since the current data and the comparison data are constituted by distance data, the obtained difference value indicates “distance difference”. For example, when the background data 200 and the object 201 are included in the current data and only the background 200 is included in the comparison data, the obtained difference value indicates “distance between the background and the object”.

  Since the difference value obtained in step ST5 is 8-bit multi-value data, the first change area extraction unit 40 determines whether or not the obtained difference value is equal to or greater than a preset threshold value (step ST6). . If the difference value is greater than or equal to the threshold value (step ST6; YES), the pixel area is extracted as a change area (step ST7). On the other hand, if the difference value is less than the threshold value (step ST6; NO), it is determined that the pixel area is not a change area (step ST8), and the process proceeds to step ST9. Thereafter, the first change area extraction unit 40 determines whether or not processing has been performed for all 80 × 60 pixels (step ST9). When the process is not performed for all 80 × 60 pixels (step ST9; NO), the process returns to step ST5 and the above-described process is repeated.

  On the other hand, when processing has been performed for all 80 × 60 pixels (step ST9; YES), the recognition processing unit 50 determines whether or not the change region extracted in step ST7 satisfies the collation condition (step ST10). If the verification condition is satisfied (step ST10; YES), the change area is recognized as a notification target (step ST11). On the other hand, when the verification condition is not satisfied (step ST10; NO), it is determined that the change area is not a notification target (step ST12), and the process returns to step ST1. The notification processing unit 60 performs notification processing on the notification target recognized in step ST11 (step ST13), and returns to the processing of step ST1.

Next, details of the determination processing by the recognition processing unit 50 will be described.
FIG. 5 is a flowchart illustrating a determination process of the recognition processing unit of the monitoring apparatus according to the first embodiment.
The recognition processing unit 50 determines whether or not the change area exists within the monitoring range (step ST21). When it exists in the monitoring range (step ST21; YES), it is further determined whether or not the change area has a predetermined area (step ST22). When it has a predetermined area (step ST22; YES), it is further determined whether or not the change area has a predetermined vertical and horizontal dimension (step ST23). If it has predetermined vertical and horizontal dimensions (step ST23; YES), it is further determined whether or not the change area has a predetermined moving speed (step ST24). When it has a predetermined moving speed (step ST24; YES), it progresses to step ST11 and it recognizes that a change area | region is a notification object.

  On the other hand, the change area does not exist within the monitoring range (step ST21; NO), does not have a predetermined area (step ST22; NO), or does not have a predetermined vertical / horizontal dimension (step ST23; NO), when it does not have a predetermined movement speed (step ST24; NO), it progresses to step ST12 and it is judged that it is not information object.

  As described above, according to the first embodiment, the current data calculation unit 20 that acquires current data from the distance data acquired by the three-dimensional laser scanner 10 and the distance data acquired by the three-dimensional laser scanner 10 are used. Since it comprised so that the comparison data calculating part 30 which acquires comparison data, and the 1st change area extraction part 40 which extracts a change area from the difference value of present data and comparison data may be provided as a difference value Information on the distance to the background can be obtained, and the difference between the object and the background can be recognized even when there is no sufficient luminance difference between the object and the background. Further, by using the Z-axis information included in the distance information acquired by the three-dimensional laser scanner 10, it is possible to recognize the movement of the object in the Z-axis direction on the three-dimensional coordinates. As a result, the recognition accuracy of the object can be improved.

Embodiment 2. FIG.
FIG. 6 is a block diagram illustrating a configuration of the monitoring apparatus according to the second embodiment.
The monitoring device 100a according to the second embodiment is provided with a positive / negative map matching unit 41 in addition to the first change region extraction unit 40 of the monitoring device 100 according to the first embodiment shown in FIG. In the following, the same or corresponding parts as those of the monitoring apparatus 100 according to the first embodiment are denoted by the same reference numerals as those used in the first embodiment, and the description thereof is omitted or simplified.

  As in the first embodiment, the first change area extraction unit 40 acquires 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 them with the current data. The difference value is calculated by comparing the data with the pixel unit. The positive / negative map collating unit 41 refers to whether the difference value calculated by the first change region extracting unit 40 is a positive value or a negative value, and collates with map data set in advance. The sign of the difference value is based on the distance relationship between the background 200 and the object 201. When the object 201 is in front of the background 200, the difference value is “positive” and the background 200 is the object 201. If it is in front, the difference value is “negative”.

  Usually, when the background 200 is a solid shielding object such as a wall, the object 201 always passes the near side of the wall. An object passing through the back side of the wall cannot be visually recognized and is not scanned by the three-dimensional laser scanner 10. On the other hand, when the background 200 is, for example, a net fence, the object 201 is scanned by the three-dimensional laser scanner 10 that can be visually recognized regardless of whether the object 201 exists on the front side or the back side of the fence.

  Thus, depending on the conditions of the background 200, it is necessary to limit the difference value based on the distance relationship between the background 200 and the object 201, that is, the sign of the difference value. The positive / negative map collation unit 41 stores in advance conditions for limiting the difference value as map data, collates the map data with the difference value, and corrects the difference value rejected by the map data condition.

  Specifically, when the map data of the pixel (x, y) is “0”, the condition of the map data is “only the positive value of the pixel difference value of the pixel (x, y) is allowed”. Become. Under this condition, when a negative value is calculated as the difference value of the pixel (x, y), the pixel is regarded as noise, and the difference value is forcibly corrected to “0”. This eliminates a difference value that does not occur depending on the conditions of the background 200, and improves the accuracy of the calculated difference value. The first change region extraction unit 40 determines whether or not the difference value collated by the positive / negative map collation unit 41 is equal to or greater than a preset threshold value, and extracts a pixel region that is equal to or greater than the threshold value as a change region.

Next, the operation of the monitoring apparatus 100a according to Embodiment 2 will be described.
FIG. 7 is a flowchart showing the operation of the monitoring apparatus according to the second embodiment.
In the following, the same steps as those of the monitoring apparatus 100 according to the first embodiment are denoted by the same reference numerals as those used in FIG. 4, and the description thereof is omitted or simplified.
In step ST5, when the first change area extraction unit 40 calculates the difference value for each pixel using the current data stored in the current data storage unit 21 and the comparison data stored in the comparison data storage unit 31. The positive / negative map collating unit 41 collates the calculated difference value with the map data (step ST31), and determines whether or not the map data is rejected (step ST32).

  When the map data condition is rejected (step ST32; YES), the positive / negative map matching unit 41 corrects the difference value (step ST33). On the other hand, if it is not rejected by the map data condition (step ST32; NO), the process proceeds to step ST6. The first change region extraction unit 40 determines whether or not the difference value obtained in step ST5 or step ST33 is greater than or equal to a preset threshold value (step ST6). Thereafter, the flowchart performs the processing from step ST7 to step ST13 described above.

  As described above, according to the second embodiment, the difference value calculated by the first change region extraction unit 40 is collated with the map data in which the condition is set in advance, and the map data condition is rejected. Therefore, the difference value that cannot be generated depending on the background condition can be eliminated, and the accuracy of the calculated difference value can be improved. Thereby, the recognition accuracy of the object can be improved.

Embodiment 3 FIG.
FIG. 8 is a block diagram illustrating a configuration of the monitoring apparatus according to the third embodiment.
In the following, the same or corresponding parts as those of the monitoring apparatus 100 according to the first embodiment are denoted by the same reference numerals as those used in the first embodiment, and the description thereof is omitted or simplified.
The monitoring apparatus 100b according to the third embodiment includes a three-dimensional laser scanner 10, a distance histogram creation unit 71, an object distance peak point calculation unit 72, a background distance peak point calculation unit 73, and a distance difference calculation unit 74. The distance calculation unit 70, the distance difference connection processing unit 75, the second change area extraction unit 42, the recognition processing unit 50, and the notification processing unit 60. In FIG. 8, the background 200 indicating the scan range of the three-dimensional laser scanner 10, the object 201 standing in front of the background 200, and the diffusion region 202 of the laser light pulse 12 are illustrated outside the monitoring apparatus 100 b. The purpose of the monitoring apparatus 100b is to recognize and notify the object 201.

  The three-dimensional laser scanner 10 includes the laser light emitting unit 11, the dispersion mechanism 13, and the laser light receiving unit 16 shown in FIG. 2 as in the first embodiment, and scans the range indicated by the background 200 to obtain distance data and intensity data. To get. The laser light pulse 12 emitted from the three-dimensional laser scanner 10 divides the field of view into areas corresponding to the resolution of the three-dimensional laser scanner 10 and irradiates one pulse to the divided area.

  The laser light pulse 12 has a characteristic of diffusing. When the background 200 shown in FIG. 8 is a flat wall, for example, the laser light pulse 12 diffuses into the diffusion region 202 on the wall surface. The diffusion region 202 indicates the diffusion range of the laser light pulse 12, and the diffusion range is determined by the distance from the three-dimensional laser scanner 10 to the background 200. For example, when the distance from the three-dimensional laser scanner 10 to the background 200 is 100 m, the diameter 202a of the diffusion region 202 is 30 cm. The laser light pulse 12 diffused in the diffusion region 202 is reflected by the background 200 or the object 201 and enters the three-dimensional laser scanner 10 as the laser reflected light 15. The three-dimensional laser scanner 10 calculates the distance and reflectance to the background 200 or the object 201 based on the received laser reflected light 15, and acquires distance data and intensity data. The acquired distance data is output to the distance histogram creation unit 71.

  The distance histogram creation unit 71 creates a distance histogram for each pixel based on the distance data input from the three-dimensional laser scanner 10. More specifically, the distance histogram creation unit 71 creates a first distance histogram for the pixel (x, y) based on the distance data for the past 10 scans of the three-dimensional laser scanner 10. The first distance histogram is created for all pixels. Further, a second distance histogram at the pixel (x, y) is created based on distance data for the past 50 scans of the three-dimensional laser scanner 10. The creation of the second distance histogram is performed for all pixels.

FIG. 9 is a diagram illustrating an example of the process of the distance calculation unit of the monitoring apparatus according to the third embodiment, and FIG. 9A illustrates an example of the distance histogram created by the distance histogram creation unit 71.
In FIG. 9A, the horizontal axis indicates the distance (m), and the vertical axis indicates the frequency. The distance histogram of the pixel (x, y) shown in FIG. 9A is a distance histogram of coordinates where the background 200 that is 100 m away from the three-dimensional laser scanner 10 and the object 201 overlap the near side of the background 200. A high frequency in the vicinity of the distance of 100 m indicates a wall that is the background 200, a high frequency in the vicinity of the distance of 30 m indicates the object 201, and the other distributions are errors due to distance jitter.

  Here, the distance jitter will be described. The pixel (x, y) described above is a coordinate at which the object 201 overlaps the near side of the background 200, and the background 200 and the object 201 are both irradiated with the laser light pulse 12. As a result, the distance data that is the calculation result of the three-dimensional laser scanner 10 fluctuates to the distance of the background 200 when the ratio of the background 200 is high, and fluctuates the distance of the object 201 when the ratio of the object 201 is high. These distance fluctuations are distance jitter. When distance jitter has occurred, the distance to the object 201 fluctuates, so that it is difficult to acquire stable distance data, and the accuracy of extraction of the change region is reduced. However, the configuration of the third embodiment Is applied to suppress errors due to distance jitter and improve the extraction accuracy of the change region.

The object distance peak point calculation unit 72 calculates a peak point by smoothing the first distance histogram created by the distance histogram creation unit 71. FIG. 9B is a diagram illustrating calculation of the object distance peak point at the pixel (x, y).
As an example of calculating the peak point, the average of the frequencies at the distance P ± 10 is calculated for the frequency at a certain distance P and replaced with the frequency at the distance P (processing 1a). This replacement process is performed for all distances 0 m to 150 m in the distance histogram. If the replaced frequencies are connected by a line, a gentle curve is obtained (Process 2a). The slope of the obtained curve is measured from a position at a distance of 0 m, and the point at which the slope changes from a right-up slope to a right-down slope is calculated as a peak point (processing 3a, arrows a and b in FIG. 9B). , C). This peak point calculation process is performed up to a distance of 150 m (process 4a). The obtained peak point becomes the target object distance peak point.

  In some cases, a plurality of peak points are calculated. For example, when three or more peak points are calculated, processing for reducing the peak points is performed. As processing for reducing the peak points, the above-described processing 1a to processing 4a are repeated until the peak points converge to about three. Thereby, a histogram curve including about three peak points is obtained. The peak point existing in the histogram curve is the target object distance peak point.

  Next, the background distance peak point calculation unit 73 calculates a peak point by smoothing the second distance histogram. FIG. 9C is a diagram showing calculation of the background distance peak point at the pixel (x, y). As an example of calculating the peak point, the average of the frequency of the distance P ± 10 is calculated with respect to the frequency at a certain distance P and replaced with the frequency of the distance P (processing 1b). This replacement process is performed for all distances from 0 m to 150 m in the histogram. If the replaced frequencies are connected by a line, a gentle curve is obtained (Process 2b). The slope of the obtained curve is measured from a position at a distance of 0 m, and a point at which the slope moves from a right-up slope to a right-down slope is calculated as a peak point (processing 3b). This peak point calculation process is performed up to a distance of 150 m (process 4b). The obtained peak point becomes the target background distance peak point.

  In some cases, a plurality of peak points are calculated. For example, when two or more peak points are calculated, processing for reducing the peak points is performed. As processing for reducing the peak points, the above-described processing 1b to processing 4b are repeated until the peak points converge to one. As a result, a histogram curve including one peak point is obtained. The peak point existing in the histogram curve is the target background distance peak point.

  The difference between the first distance histogram and the second distance histogram is the time required to create the histogram. As described above, the first distance histogram is created based on the distance data for the past 10 scans of the three-dimensional laser scanner 10, and the second distance histogram is the distance data for the past 50 scans of the three-dimensional laser scanner 10. Create based on. Thus, the second distance histogram uses five times as much time as the first distance histogram, and the second distance histogram is insensitive to short-time changes compared to the first distance histogram. (Indicating a long-term change in the monitoring area) and a histogram that is strongly influenced by the distance of the fixed background. On the other hand, the first distance histogram is sensitive to a short-time change (indicating a short-time change in the monitoring area) and is a histogram that is strongly influenced by the distance of the object. Thus, the object distance peak point and the background distance peak point are calculated.

The distance difference calculation unit 74 calculates a difference from a histogram curve including an object distance peak point (hereinafter referred to as an object distance histogram curve) and a histogram curve including a background distance peak point (hereinafter referred to as a background distance histogram curve). Calculate the peak point. FIG. 9D is a diagram illustrating difference peak point calculation at the pixel (x, y).
As a specific calculation method, first, the background distance histogram curve shown in FIG. 9C is regarded as a normal distribution curve, and the standard deviation (distance from the mean μ to the inflection point) σ of the normal distribution is used. A range 3σ _a centering on the background distance peak point C (average μ in the normal distribution) is determined (processing 1c). Then, the subject distance histogram curve, deletes the peak point located in the range 3 [sigma] _a (processing 2c). The histogram 2 shown in FIG. 9D is obtained by the process 2c. The maximum peak point included in the obtained histogram curve is calculated as the difference peak point D (processing 3c). Finally, the histogram including the difference peak point D is regarded as a normal distribution curve, the range 3σ _b centering on the difference peak point D is determined, and the position of the difference peak point D and the position of the range 3σ _b are stored (processing 4c). ).

  Next, extraction of a change region based on the difference peak point will be described with reference to FIG. FIG. 10 is a diagram illustrating extraction of a change region of the monitoring apparatus according to the third embodiment. FIG. 10A and FIG. 10B are explanatory views showing creation of an integrated region by the distance difference connection processing unit, and FIG. 10A is a diagram in which the object 201 is arranged on the background 200, and FIG. ) Is a diagram showing only the background 200. FIG. 10C is a diagram illustrating the change region extracted by the second change region extraction unit.

  The distance difference connection processing unit 75 connects the distance information indicated by the difference peak points of all the pixels calculated by the distance difference calculation unit 74 to create an integrated region. For example, when the difference peak point is calculated for each pixel of 80 × 60 pixels, the distance difference connection processing unit 75 places the distance information indicated by the difference peak point in the virtually created 80 × 60 pixel cell. To create an integrated area. As shown in FIGS. 10 (a) and 10 (b), the disposition of distance information is, for example, a pixel where a difference peak point exists is colored gray, and a pixel where no difference peak point exists is colored white. Make visual arrangements. In FIG. 10A, the position where the object 201 exists and the pixel position of the integrated region 203 coincide. Note that the coloring of the pixels is not limited to gray and can be changed as appropriate.

  However, since there is an error in the difference peak point calculation processing by the distance difference calculation unit 74, a pixel colored in gray occurs at a position where the object 201 does not exist, or is colored gray at a position where the object 201 should exist. There may be a case where no pixel is generated. Therefore, in the second change region extraction unit 42, filter processing is executed on the integrated region created by the distance difference connection processing unit 75, and the change region due to noise is deleted to change only the change region corresponding to the object 201. To extract. Specifically, referring to the position of the difference peak point calculated by the distance difference calculation unit 74 and the position of the range 3σ, it is similar to at least 3 pixels among the surrounding 8 pixels centered on a certain pixel (x, y) It is determined whether or not there is a difference peak point. Whether there is a similar difference peak point is determined by whether another difference peak point exists in the range 3σ around the difference peak point of a certain pixel (x, y). This process is performed for all pixels.

  If there is a differential peak similar to at least three pixels centered on the pixel (x, y), it is determined that the pixel (x, y) is a change area of the object 201 to be extracted. On the other hand, when there is no difference peak similar to at least three pixels centering on the pixel (x, y), the pixel (x, y) is determined to be a change region due to noise, and the pixel (x, y) is determined. ) Delete the difference peak information. After performing the above-described filtering process for all pixels, the area determined to be the change area of the object 201 is re-arranged in the 80 × 60 pixel cell that is virtually created again to determine the final change area. To do. FIG. 10C shows the change region 204 after the filter processing by the second change region extraction unit 42.

  The recognition processing unit 50 determines that the change region is a notification target based on whether or not the condition of the change region extracted by the second change region extraction unit 42 satisfies a predetermined condition. The notification processing unit 60 performs notification processing based on the recognition result of the recognition processing unit 50. Here, the notification process is a process of transmitting a specific signal to a higher-level PC or the like, or a process of sounding a buzzer of the apparatus.

Next, the operation of the monitoring apparatus 100b according to the third embodiment will be described.
FIG. 11 is a flowchart showing the operation of the monitoring apparatus according to the third embodiment.
In the following, the same steps as those of the monitoring apparatus 100 according to the first embodiment are denoted by the same reference numerals as those used in FIG. 4, and the description thereof is omitted or simplified. Further, as in the first embodiment, the case where the resolution of the three-dimensional laser scanner 10 is 80 × 60 pixels will be described as an example.
First, the three-dimensional laser scanner 10 scans the range of the background 200 (step ST1), and acquires distance data and intensity data (step ST2). Specifically, the range of the background 200 is divided into 80 × 60, which is the resolution of the three-dimensional laser scanner 10, and scanned. The distance data and the intensity data are generally digital data, and here, multi-value data of 8 bits per pixel of 80 × 60 pixels is used.

  The distance histogram creation unit 71 creates a first distance histogram and a second distance histogram based on the distance data acquired in step ST2 (step ST41). The object distance peak point calculation unit 72 calculates an object distance peak point from the first distance histogram created in step ST41 (step ST42), and creates an object distance histogram curve (step ST43). The background distance peak point calculation unit 73 calculates a background distance peak point from the second distance histogram created in step ST41 (step ST44), and creates a background distance histogram curve (step ST45).

  Next, the distance difference calculation unit 74 calculates a difference peak point between the object distance histogram curve created in step ST43 and the background distance histogram curve created in step ST45 (step ST46). The distance calculation unit 70 determines whether or not the processing from step ST42 to step 46 has been performed for all the pixels (step ST47). If all the pixels have not been processed (step ST47; NO), the process returns to step ST41 and the above-described processing is repeated.

  On the other hand, when the processing is performed for all the pixels (step ST47; YES), the distance difference connection processing unit 75 connects the distance information of the difference peak points of all the pixels obtained in step ST46, and creates an integrated region (step). ST48). The second change area extraction unit 42 performs a filtering process on the integrated area created in step ST48 to extract a change area (step ST49). The recognition processing unit 50 determines whether or not the change area extracted in step ST49 satisfies the verification condition (step ST50). When the verification condition is satisfied (step ST50; YES), the change area is recognized as a notification target (step ST11). On the other hand, if the verification condition is not satisfied (step ST50; NO), it is determined that the change area is not a notification target (step ST12), and the process returns to step ST1. The notification processing unit 60 performs notification processing on the notification target recognized in step ST11 (step ST13), and returns to the processing of step ST1.

Next, details of the determination processing of the recognition processing unit 50 will be described.
FIG. 12 is a flowchart illustrating the determination process of the recognition processing unit of the monitoring apparatus according to the third embodiment.
The recognition processing unit 50 determines whether or not the change area has a predetermined area (step ST51). When it has a predetermined area (step ST51; YES), it is further determined whether or not the change area has a predetermined vertical and horizontal dimension (step ST52). If it has predetermined vertical and horizontal dimensions (step ST52; YES), it is further determined whether or not the change area has a predetermined moving speed (step ST53). If it has a predetermined moving speed (step ST53; YES), it is further determined whether or not the change area has a predetermined existence time (step ST54). When it has the predetermined existence time (step ST54; YES), the process proceeds to step ST11, and the change area is recognized as the notification target.

  On the other hand, when it does not have a predetermined area (step ST51; NO), when it does not have a predetermined vertical and horizontal dimension (step ST52; NO), when it does not have a predetermined moving speed (step ST53; NO) ), When it does not have the predetermined existence time (step ST54; NO), the process proceeds to step ST12, and it is determined that the change region is not a notification target.

As described above, according to the third embodiment, the distance histogram creating unit 71 that creates the distance histogram from the distance data acquired by the three-dimensional laser scanner 10 and the object including the object distance peak point from the distance histogram. An object distance peak point calculation unit 72 that creates a distance histogram curve, a background distance peak point calculation unit 73 that creates a background distance histogram curve including a background distance peak point from the distance histogram, an object distance histogram curve and a background distance histogram A difference difference calculation unit 74 that calculates a difference peak point from a curve, a distance difference connection processing unit 75 that creates an integrated region using distance information of the difference peak point, and a filter process for the integrated region, and a change region And a second change region extraction unit 42 for extracting the background, the object, Distance-difference peak point can be obtained based on the relationship, it is possible to recognize the difference also between the object and the background when a sufficient difference in brightness and the object and the background is not present. Thereby, the recognition accuracy of the object can be improved.
In addition, it is possible to extract only the change area of the object while suppressing the distance jitter (distance fluctuation) generated at the coordinates where the background and the object overlap, and improve the recognition accuracy of the object.

Embodiment 4 FIG.
FIG. 13 is a block diagram illustrating a configuration of the monitoring apparatus according to the fourth embodiment.
The monitoring device 100c according to the fourth embodiment includes a background distance peak information accumulation unit 76 instead of the background distance peak point calculation unit 73 of the monitoring device 100b according to the third embodiment shown in FIG. In the following description, the same or corresponding parts as those of the monitoring apparatus 100b according to the third embodiment are denoted by the same reference numerals as those used in the third embodiment, and the description thereof is omitted or simplified. In FIG. 13, the background 200 indicating the scan range of the three-dimensional laser scanner 10, the object 201 standing in front of the background 200, and the diffusion region 202 of the laser light pulse 12 are illustrated outside the monitoring apparatus 100 c.

The distance histogram creation unit 71a creates a distance histogram for each pixel based on the distance data input from the three-dimensional laser scanner 10. However, based on the distance data for the past 10 scans of the three-dimensional laser scanner 10, only the first distance histogram for the pixel (x, y) is created. The background distance peak information accumulating unit 76 stores background distance peak points based on a histogram (corresponding to the second distance histogram shown in the third embodiment described above) that is strongly influenced by the distance to the fixedly existing background 200. The position information and the background distance peak information in which the position information of the range 3σ _a centered on the background distance peak point is fixed in advance are stored.

The distance difference calculation unit 74 a acquires the histogram curve including the object distance peak point calculated by the object distance peak point calculation unit 72 (hereinafter referred to as the object distance histogram curve) and the background distance peak information accumulation unit 76. A difference peak point is calculated from the background distance peak information. Specifically, with reference to FIG. 9 of the third embodiment described above, a histogram curve including a difference peak point by deleting a peak point located in the range 3σ _a which is a fixed value from the object distance histogram curve. (See FIGS. 9C and 9D). The maximum peak point included in the obtained histogram curve is calculated as a difference peak point, and a range 3σ _b around the difference peak point is determined (see FIG. 9D). The position of the obtained difference peak point and the position of the range 3σ _b centered on the difference peak point are stored.

Next, the operation of the monitoring device 100c according to Embodiment 4 will be described.
FIG. 14 is a flowchart showing the operation of the monitoring apparatus according to the fourth embodiment.
In the following, the same steps as those of the monitoring device 100b according to the third embodiment are denoted by the same reference numerals as those used in FIG. 11, and the description thereof is omitted or simplified.
The distance histogram creation unit 71a creates a first distance histogram based on the distance data acquired in step ST2 (step ST61). The object distance peak point calculation unit 72 calculates an object distance peak point from the first distance histogram created in step ST61 (step ST42), and creates an object distance histogram curve (step ST43).

  The distance difference calculation unit 74a acquires background distance peak information from the background distance peak information storage unit 76 (step ST62), the object distance histogram curve created in step ST43, and the background distance peak information acquired in step ST62. The difference peak point is calculated from (step ST63). The distance calculation unit 70a determines whether or not the processing from step ST61 to step 63 has been performed for all the pixels (step ST47). When the process is not performed for all the pixels (step ST47; NO), the process returns to step ST61 and the above-described process is repeated. On the other hand, if all pixels have been processed (step ST47; YES), the process proceeds to step ST48.

As described above, according to the fourth embodiment, the position information of the background distance peak point based on the histogram that is strongly influenced by the distance to the fixedly existing background 200 and the range centered on the background distance peak point. Since the configuration is provided with the background distance peak information accumulating unit 76 for storing the background distance peak information in which the position information of 3σ _a is fixed in advance, the process of creating the second histogram and the background distance peak point are calculated. It is possible to reduce the processing load. Even when the object stays in one place and does not move, the background can be recognized from the predetermined background distance peak information, and the object that does not move can be distinguished from the background.

Embodiment 5. FIG.
In the above-described third embodiment, the configuration in which the change area is extracted based on the distance data of the three-dimensional laser scanner 10 has been described. However, in the fifth embodiment, the change area is determined based on the intensity data of the three-dimensional laser scanner 10. The structure to extract is shown.
The resolution of the distance of the laser light pulse 12 has a limit. For example, when the distance resolution of the laser light pulse 12 is 50 cm and the background 200 and the object 201 are at a distance of 50 cm or less, the background 200 and the object 201 cannot be distinguished. This is a problem in distance resolution, which occurs when the object 201 moves along the background 200 such as a wall, and the object 201 cannot be found in distance resolution. The fifth embodiment shows a configuration in which the object 201 that cannot be recognized in terms of distance resolution can be recognized using the intensity data input from the three-dimensional laser scanner 10.

FIG. 15 is a block diagram illustrating a configuration of the monitoring apparatus according to the fifth embodiment.
In the following description, the same or corresponding parts as those of the monitoring apparatus 100b according to the third embodiment are denoted by the same reference numerals as those used in the third embodiment, and the description thereof is omitted or simplified.
The monitoring apparatus 100d according to the fifth embodiment includes a three-dimensional laser scanner 10, an intensity histogram creation unit 81, an object intensity peak point calculation unit 82, a background intensity peak point calculation unit 83, and an intensity difference calculation unit 84. The intensity calculation unit 80, the intensity difference connection processing unit 85, the third change area extraction unit 86, the recognition processing unit 50, and the notification processing unit 60 are included. In FIG. 15, the background 200 indicating the scan range of the three-dimensional laser scanner 10, the object 201 standing in front of the background 200, and the diffusion region 202 of the laser light pulse 12 are shown outside the monitoring apparatus 100 d.

  The configuration of the three-dimensional laser scanner 10 is the same as that of the first to fourth embodiments. The intensity histogram creation unit 81 creates an intensity histogram for each pixel based on the intensity data input from the three-dimensional laser scanner 10. More specifically, the intensity histogram creating unit 81 creates a first intensity histogram for the pixel (x, y) based on intensity data for the past 10 scans of the three-dimensional laser scanner 10. The first intensity histogram is created for all pixels. Further, a second intensity histogram for the pixel (x, y) is created based on the intensity data for the past 50 scans, and the second intensity histogram is created for all the pixels.

FIG. 16 is a diagram illustrating an example of processing of the intensity calculation unit of the monitoring apparatus according to the fifth embodiment, and FIG. 16A illustrates an example of a histogram created by the intensity histogram creation unit 81.
In FIG. 16A, the horizontal axis indicates intensity (%), and the vertical axis indicates frequency. In the intensity histogram of the pixel (x, y) shown in FIG. 16A, the frequency around the intensity of 20% is high, and the reflectance of the object 201 is reflected. The other frequency is noise.

  Next, the object intensity peak point calculation unit 82 calculates the peak point by smoothing the first intensity histogram. FIG. 16B is a diagram illustrating calculation of the object intensity peak point at the pixel (x, y). As an example of calculating the peak point, the average of the frequencies of the intensity Q ± 10 is calculated for the frequency at a certain intensity Q and replaced with the frequency of the intensity Q (processing 1d). This replacement process is performed for all of the intensity histograms having an intensity of 0% to 100%. If the replaced frequencies are connected by a line, a gentle curve is obtained (Process 2d). A slope is measured from the position where the intensity is 0% with respect to the obtained curve, and a point at which the slope moves from a right-up slope to a right-down slope is calculated as a peak point (Process 3d, arrow d, FIG. 16B). e, f). This peak point calculation process is performed up to the position where the intensity is 100% (process 4d). The obtained peak point becomes the target object strength peak point.

  In some cases, a plurality of peak points are calculated. For example, when three or more peak points are calculated, processing for reducing the peak points is performed. As processing for reducing the peak points, the above-described processing 1d to processing 4d are repeated until the peak points converge to about three. Thereby, a histogram curve including about three peak points is obtained. The peak point existing in the histogram curve is the target object intensity peak point.

Next, the background intensity peak point calculation unit 83 calculates the peak point by smoothing the second intensity histogram. FIG. 16C is a diagram showing calculation of the background intensity peak point at the pixel (x, y).
As an example of calculating the peak point, the average of the frequencies of the intensity Q ± 10 is calculated with respect to the frequency at a certain intensity Q and replaced with the frequency of the intensity Q (processing 1e). This replacement process is performed for all the histograms having an intensity of 0% to 100%. When the replaced frequencies are connected with a line, a gentle curve is obtained (Process 2e). The inclination is measured from the position where the intensity is 0% with respect to the obtained curve, and the point at which the slope moves from the upward slope to the downward slope is calculated as a peak point (processing 3e). This peak point calculation process is performed up to the position where the intensity is 100% (process 4e). The obtained peak point becomes the target background intensity peak point.

  In some cases, a plurality of peak points are calculated. For example, when two or more peak points are calculated, processing for reducing the peak points is performed. As processing for reducing the peak points, the above-described processing 1e to processing 4e are repeated until the peak points converge to one. As a result, a histogram curve including one peak point is obtained. The peak point existing in the histogram curve is the target background intensity peak point.

  The difference between the first intensity histogram and the second intensity histogram is the time required to create the histogram. As described above, the first intensity histogram is created based on the intensity data for the past 10 scans of the three-dimensional laser scanner 10, and the second intensity histogram is the intensity data for the past 50 scans of the three-dimensional laser scanner 10. Create based on. Thus, the second intensity histogram uses five times as much time as the first intensity histogram, and the second intensity histogram is insensitive to short-time changes compared to the first intensity histogram. (Indicating a long-term change in the monitoring area), and a histogram that is strongly influenced by the intensity of the background that exists fixedly. On the other hand, the first intensity histogram is sensitive to a short-time change (indicating a short-time change in the monitoring area), and is a histogram that is strongly influenced by the intensity of the object. Thus, the object intensity peak point and the background intensity peak point are calculated.

The intensity difference calculation unit 84 calculates a difference from a histogram curve including an object intensity peak point (hereinafter referred to as an object intensity histogram curve) and a histogram curve including a background intensity peak point (hereinafter referred to as a background intensity histogram curve). Calculate the peak point. FIG. 16D is a diagram illustrating calculation of a difference peak point at the pixel (x, y).
As a specific calculation method, first, the background intensity histogram curve shown in FIG. 16C is regarded as a normal distribution curve, and the standard deviation (intensity from the mean μ to the inflection point) σ of the normal distribution is used. A range 3σ _c centered on the intensity peak point F (average μ in the normal distribution) is determined (processing 1f). Next, the peak point located within the range 3σ _c is deleted from the object intensity histogram curve (processing 2f). The histogram 2 shown in FIG. 16D is obtained by the process 2f. The maximum peak point included in the obtained histogram curve is calculated as the difference peak point G (processing 3f). Finally, the histogram including the difference peak point G is regarded as a normal distribution curve, the range 3σ _d centered on the difference peak point G is determined, and the position of the difference peak point G and the position of the range 3σ _d are stored (processing 4f). ).

  Next, extraction of a change area based on the difference peak point will be described with reference to FIG. FIG. 17 is a diagram illustrating extraction of a change area of the monitoring apparatus according to the fifth embodiment. FIG. 17A and FIG. 17B are explanatory views showing extraction of a change region by the intensity difference connection processing unit and the third change region extraction unit, and FIG. 17A shows the object 201 on the background 200. The arranged diagram and FIG. 17B are diagrams showing only the background 200.

  The intensity difference connection processing unit 85 connects the intensity information indicated by the difference peak points of all the pixels calculated by the intensity difference calculation unit 84. For example, when a difference peak point is calculated for each pixel of 80 × 60 pixels, the intensity difference connection processing unit 85 places the intensity information indicated by the difference peak point in a virtually created 80 × 60 pixel cell. Connect. For example, as shown in FIGS. 17A and 17B, the arrangement of the intensity information is such that a pixel having a difference peak point is colored gray and a pixel having no difference peak point is colored white. And so on.

  The third change region extraction unit 86 extracts a change region from the intensity information connected by the intensity difference connection processing unit 85. The change area is extracted based on the difference between the pixel where the difference peak point exists and the pixel where the difference peak point does not exist, and in the above-described example, the pixel colored gray based on the difference in pixel coloring is extracted. The gathering area is extracted as a change area. As shown in FIG. 17A, the position where the object 201 exists and the pixel position of the change area 205 coincide.

  The recognition processing unit 50 recognizes whether or not the change region is a notification target based on whether or not the condition of the change region extracted by the third change region extraction unit 86 satisfies a predetermined condition. . The notification processing unit 60 performs notification processing based on the recognition result of the recognition processing unit 50. Here, the notification process is a process of transmitting a specific signal to a higher-level PC or the like, or a process of sounding a buzzer of the apparatus.

Next, the operation of the monitoring device 100d according to the fifth embodiment will be described.
FIG. 18 is a flowchart showing the operation of the monitoring apparatus according to the fifth embodiment.
In the following, the same steps as those of the monitoring device 100b according to the third embodiment are denoted by the same reference numerals as those used in FIG. 11, and the description thereof is omitted or simplified. Further, as in the third embodiment, the case where the resolution of the three-dimensional laser scanner 10 is 80 × 60 pixels will be described as an example.

  The intensity histogram creation unit 81 creates a first intensity histogram and a second intensity histogram based on the intensity data acquired in step ST2 (step ST71). The object intensity peak point calculation unit 82 calculates an object intensity peak point from the first intensity histogram created in step ST71 (step ST72), and creates an object intensity histogram curve (step ST73). The background intensity peak point calculation unit 83 calculates a peak point from the second intensity histogram created in step ST72 (step ST74), and creates a background intensity histogram curve (step ST75).

  Next, the intensity difference calculation unit 84 calculates a difference peak point between the object intensity histogram curve created in step ST73 and the background intensity histogram curve created in step ST75 (step ST76). The intensity calculation unit 80 determines whether or not the processing from step ST71 to step ST76 has been performed for all pixels (step ST77). If the process has not been performed for all pixels (step ST77; NO), the process returns to step ST71 and the above-described process is repeated.

  On the other hand, when processing has been performed for all the pixels (step ST77; YES), the intensity difference connection processing unit 85 connects the intensity information of the difference peak points of all the pixels obtained in step ST76 (step ST78). The change area extraction unit 86 extracts a change area from the connected intensity information (step ST79). The recognition processing unit 50 determines whether or not the change area extracted in step ST79 satisfies the matching condition (step ST80). If the verification condition is satisfied (step ST80; YES), the change area is recognized as a notification target (step ST11). On the other hand, when the verification condition is not satisfied (step ST80; NO), it is determined that the change area is not a notification target (step ST12), and the process returns to step ST1. The notification processing unit 60 performs notification processing on the notification target recognized in step ST11 (step ST13), and returns to the processing of step ST1.

  Details of the determination processing of the recognition processing unit 50 are the same as those in the flowchart shown in FIG.

As described above, according to the fifth embodiment, the intensity histogram creating unit 81 that creates the intensity histogram from the intensity data acquired by the three-dimensional laser scanner 10 and the object including the object intensity peak point from the intensity histogram. An object intensity peak point calculation unit 82 that creates an intensity histogram curve, a background intensity peak point calculation unit 83 that creates a background intensity histogram curve including a background intensity peak point from the intensity histogram, an object intensity histogram curve, and a background intensity histogram An intensity difference calculation unit 84 that calculates a difference peak point from the curve, an intensity difference connection processing unit 85 that connects intensity information of the difference peak points, and a third change area extraction that extracts a change area from the connected intensity information Difference peak based on the intensity relationship between the background and the object. Can be obtained, it is possible to recognize the difference also between the object and the background when a sufficient difference in brightness and the object and the background is not present. Thereby, the recognition accuracy of the object can be improved.
Further, even when the distance between the object and the background is equal to or less than the distance resolution of the laser light pulse 12, only the change region of the object is extracted using the intensity data based on the reflectance from the background and the object. Can do. Thereby, the recognition accuracy of the object can be improved.

Embodiment 6 FIG.
In the sixth embodiment, in addition to the configuration of the fifth embodiment described above, a configuration is shown in which a target object that moves like a “person” and a target object that does not move like a “signboard” are distinguished and notified. .
FIG. 19 is a block diagram illustrating a configuration of the monitoring apparatus according to the sixth embodiment.
The monitoring apparatus 100e of the sixth embodiment is provided with a stability confirmation unit 43 in addition to the monitoring apparatus 100d of the fifth embodiment shown in FIG. In the following description, the same or corresponding parts as those of the monitoring apparatus 100d according to the fifth embodiment are denoted by the same reference numerals as those used in the fifth embodiment, and description thereof is omitted or simplified. In FIG. 19, a background 200 indicating the scanning range of the three-dimensional laser scanner 10, an object 201 standing in front of the background 200, and a diffusion region 202 of the laser light pulse 12 are illustrated outside the monitoring device 100 e.

  The stability confirmation unit 43 temporarily accumulates the change region extracted by the third change region extraction unit 86 in the accumulation unit 43a. The stability confirmation unit 43 refers to the change region stored in the storage unit 43a, compares the change regions before and after a predetermined time, and determines whether the area change of the change region is equal to or greater than a threshold value. If the area change of the change area is equal to or greater than the threshold value, it is determined that the change area is a person. On the other hand, when the area change of the change area is less than the threshold, it is determined that the change area is other than a person, for example, a signboard, a wall stain, a wall wet, a poster, or the like. The recognition processing unit 50 performs a recognition process as to whether or not only a change area determined to be a person by the stability confirmation unit 43 is a notification target.

FIG. 20 is an explanatory diagram illustrating processing of the stability confirmation unit of the monitoring apparatus according to the sixth embodiment.
20 (a), (b), and (c) show a change area when the person 206 is positioned in front of the background 200. FIGS. 20 (d), (e), and (f) show the background 200. The change area when the signboard 207 is positioned in front is shown. In FIG. 20, the region that is colored gray by the intensity difference connection processing unit 85 and extracted by the third change region extraction unit 86 is a change region.

  First, a case where the person 206 is positioned in front of the background 200 will be described. FIG. 20A shows a person 206 and a change area 208 positioned in front of the background 200. In addition, it is assumed that the person 206 moves in the direction of the arrow R. FIG. 20B shows only the change region 208 with respect to FIG. Further, FIG. 20C shows a change area 209 after a predetermined time (for example, 1 second) has elapsed since the change area 208 of FIG. 20B was extracted. Since the person 206 moves in the direction of the arrow R, when comparing FIG. 20B and FIG. 20C, the change area 208 corresponding to the person 206 moves in the direction of the arrow R and changes to the change area 209. Yes. In this way, a change area that changes after a predetermined time has elapsed is recognized as a notification target.

  Next, the case where the signboard 207 is positioned in front of the background 200 will be described. FIG. 20D shows the signboard 207 positioned in front of the background 200 and the change area 210. FIG. 20 (e) shows only the change region 210 with respect to FIG. 20 (d). Further, FIG. 20F shows a change area 211 after a predetermined time (for example, 1 second) has elapsed since the change area 210 of FIG. 20E was extracted. Since the signboard 207 does not move, there is no change in the change area 210 and the change area 211 corresponding to the signboard 207 when comparing FIG. 20E and FIG. Thus, it is recognized that a change area that remains in the same place and does not change is not a notification target.

Next, the operation of the monitoring device 100e according to Embodiment 6 will be described.
FIG. 21 is a flowchart showing the operation of the monitoring apparatus according to the sixth embodiment.
In the following, the same steps as those of the monitoring device 100d according to the fifth embodiment are denoted by the same reference numerals as those used in FIG. 18, and the description thereof is omitted or simplified.
In step ST79, when the third change region extraction unit 86 extracts the change region, the stability confirmation unit 43 temporarily stores the extracted change region in the storage unit 43a (step ST81). The stability confirmation unit 43 determines whether or not the change region before and after the lapse of a certain time has been accumulated in the accumulation unit 43a (step ST82). If not accumulated (step ST82; NO), the process returns to step ST1, and the above-described process is repeated.

  On the other hand, when accumulated (step ST82; YES), the stability confirmation unit 43 compares the change regions accumulated in the accumulation unit 43a before and after the elapse of a predetermined time, and determines whether or not the area change of the change region is equal to or greater than a threshold value. Perform (step ST83). If the area change of the change area is equal to or greater than the threshold (step ST83; YES), it is determined that the change area is a person (step ST84), and the recognition processing unit 50 determines whether the change area satisfies the matching condition. Perform (step ST85). If the verification condition is satisfied (step ST85; YES), the change area is recognized as a notification target (step ST11). On the other hand, when the verification condition is not satisfied (step ST85; NO), it is determined that the change area is not a notification target (step ST12), and the process returns to step ST1.

  On the other hand, when the area change of the change region is less than the threshold (step ST83; NO), it is determined that the change region is other than a person (step ST86), and the process proceeds to step ST12. The notification processing unit 60 performs notification processing on the notification target recognized in step ST11 (step ST13), and returns to the processing of step ST1.

Next, details of the determination processing of the recognition processing unit 50 will be described. FIG. 22 is a flowchart illustrating the determination process of the recognition processing unit of the monitoring apparatus according to the sixth embodiment. Note that the recognition processing unit 50 of the sixth embodiment performs the following processing in which the processing of step ST53 in the flowchart shown in FIG. 12 of the third embodiment is omitted.
The recognition processing unit 50 determines whether or not the change area has a predetermined area (step ST51). When it has a predetermined area (step ST51; YES), it is further determined whether or not the change area has a predetermined vertical and horizontal dimension (step ST52). If it has predetermined vertical and horizontal dimensions (step ST52; YES), it is further determined whether or not the change area has a predetermined existence time (step ST54). When it has the predetermined existence time (step ST54; YES), the process proceeds to step ST11, and the change area is recognized as the notification target.

  On the other hand, when it does not have a predetermined area (step ST51; NO), when it does not have a predetermined vertical and horizontal dimension (step ST52; NO), when it does not have a predetermined existence time (step ST54; NO) ), The process proceeds to step ST12, where it is determined that the change region is not a notification target.

  As described above, according to the sixth embodiment, it is determined whether or not there is an area change in the change area before and after the elapse of a certain time, and a stability confirmation unit that determines whether the change area is a person or not For example, a signboard, a spot on the wall, wetness on the wall, a poster, and the like can be distinguished from a person, and erroneous notification to a person other than the person can be prevented.

  In the above-described first to sixth embodiments, the configuration in which the three-dimensional laser scanner 10 is arranged in the monitoring devices 100, 100a, 100b, 100c, 100d, and 100e has been described. The distance data and the intensity data may be acquired from the external configuration.

In addition to the above, within the scope of the present invention, the present invention can be freely combined with each embodiment, modified any component of each embodiment, or omitted any component in each embodiment. Is possible.
Specifically, examples of combinations of the embodiments are shown in FIGS.
FIG. 23 is a block diagram illustrating a configuration of a monitoring device realized by combining the configurations of the first embodiment and the third embodiment.
FIG. 24 is a block diagram showing a configuration of a monitoring apparatus realized by combining the configurations of the first embodiment and the fifth embodiment.
FIG. 25 is a block diagram showing a configuration of a monitoring device realized by combining the configurations of the third embodiment and the fifth embodiment.
FIG. 26 is a block diagram illustrating a configuration of a monitoring device realized by combining the configurations of the first embodiment, the third embodiment, and the fifth embodiment.

  Although illustration is omitted, the second embodiment is applied instead of the first embodiment, the fourth embodiment is applied instead of the third embodiment, and the sixth embodiment is applied instead of the fifth embodiment. It is also possible.

Embodiment 7 FIG.
In the dispersion mechanism 13 shown in FIG. 3 of the first embodiment, the rotating mirror 13a and the rotating mirror 13c always operate in cooperation in a series of operations, and it is necessary to maintain the same relative angle. However, errors are inherent in physical operations such as driving of the rotating mirror, and errors are also included in the operations of the rotating mirror 13a and the rotating mirror 13c of the dispersion mechanism 13. As a result, the position where the incident laser light pulse 12 is finally reflected and arrives is shifted by the error included in the operation of the rotating mirror 13a and the rotating mirror 13c.

For example, assuming that an error occurring at a position on the background is “± 1 point”, the laser light pulse 12 that has reached the background is inclined in the vertical and horizontal directions with respect to the ideal arrival position that does not include the error described above. There is a possibility of deviation by ± 1 point. That is, the laser light pulse 12 reaches any one of the 9 points around the ideal arrival position.
Therefore, in the seventh embodiment, a configuration is shown in which the surrounding pixels are compared with the current data in addition to the ideal arrival position.

FIG. 27 is a block diagram illustrating a configuration of the monitoring apparatus according to the seventh embodiment.
The monitoring device 100j according to the seventh embodiment is configured by replacing the first change region extraction unit 40 of the monitoring device 100 according to the first embodiment shown in FIG. 1 with a first change region extraction unit 40a. In the following, the same or corresponding parts as those of the monitoring apparatus 100 according to the first embodiment are denoted by the same reference numerals as those used in the first embodiment, and the description thereof is omitted or simplified.

As in the first embodiment, the comparison data calculation unit 30 acquires distance data input from the three-dimensional laser scanner 10, converts it into comparison data, and stores it in the comparison data storage unit 31. In the conversion process to comparison data, for example, distance data of a frame immediately before the input distance data is obtained and used as comparison data. Here, with respect to the coordinates or pixels of the input distance data, the accumulated distance data is the distance data of points within a range set around the coordinates or pixels.
Assuming that the error occurring at the position on the background as described above is “± 1 point” and the coordinates of the input distance data are (x, y), the comparison data is centered on the coordinates (x, y). The distance data is set to coordinates moved by ± 1 point in the diagonal direction. Specifically, vertical coordinate (x, y + 1), (x, y-1), horizontal coordinate (x + 1, y), (x-1), diagonal coordinate (x + 1, y + 1), ( The distance data set in (x + 1, y-1), (x-1, y + 1), (x-1, y-1) is stored.

  The first change area extraction unit 40a acquires the current data accumulated in the current data accumulation unit 21 and all comparison data accumulated in the comparison data accumulation unit 31, and coordinates the current data and all comparison data. Alternatively, a plurality of difference values are calculated by comparison in units of pixels. The first change area extraction unit 40a sets a minimum difference value among the calculated difference values as a true difference value, and extracts a coordinate area or a pixel area in which the true difference value is equal to or greater than a preset threshold as a change area. To do. Generally, a fixed threshold value is set, and converted into binary data depending on whether or not the difference value is greater than or equal to the set threshold value.

Next, the operation of the monitoring apparatus 100j according to Embodiment 7 will be described.
FIG. 28 is a flowchart showing the operation of the monitoring apparatus according to the seventh embodiment.
In the following, the same steps as those of the monitoring apparatus 100 according to the first embodiment are denoted by the same reference numerals as those used in FIG. 4, and the description thereof is omitted or simplified. The case where the error occurring at the position on the background is “± 1 point” will be described as an example.

  In step ST4, when the comparison data is accumulated in the comparison data accumulation unit 31, the first change area extraction unit 40a acquires the coordinates (x, y) that is the current data accumulated in the current data accumulation unit 21 ( Step ST91). Furthermore, the first change area extraction unit 40a acquires comparison data centered on the coordinates (x, y) of the current data from the comparison data storage unit 31 (step ST92). Specifically, the first change area extraction unit 40a uses the coordinates (x-1, y-1), coordinates (x, y-1), coordinates (x + 1, y-1), coordinates (x −1, y), coordinates (x, y), coordinates (x + 1, y), coordinates (x−1, y + 1), coordinates (x, y + 1), and coordinates (x + 1, y + 1) are acquired.

  The first change area extraction unit 40a calculates a difference value for each coordinate using the current data acquired in step ST91 and each comparison data acquired in step ST92 (step ST93). The first change area extraction unit 40a sets the smallest difference value among the difference values calculated in step ST93 as a true difference value (step ST94). Since the true difference value obtained in step ST94 is multi-value data of 8 bits, the first change area extraction unit 40a determines whether or not the obtained true difference value is greater than or equal to a preset threshold value. (Step ST95). If the true difference value is greater than or equal to the threshold (step ST95; YES), the pixel area is extracted as a change area (step ST7). On the other hand, when the true difference value is less than the threshold value (step ST95; NO), it is determined that the pixel area is not a change area (step ST8), and the process proceeds to step ST9.

  When the error generated at the position on the background by the series of operations of the dispersion mechanism 13 is “± 1 point”, the current data of the coordinates (x, y) is vertically and horizontally inclined around the position of the coordinates (x, y). Since it is the distance data of any point within the range of ± 1 point in the direction, even if compared with the comparison data of the same coordinates (x, y) based on the coordinates (x, y) of the current data, the comparison The data is not always correct. For this reason, all the comparison data in the range where the error may occur is prepared for the current data in which an error may occur due to a series of operations of the dispersion mechanism 13, and all the prepared comparison data and the current data The difference value is calculated by comparing the data with the brute force. In the process of step ST93 described above, the nine comparison data prepared and the current data are compared with each other to calculate a difference value.

  In order to select the true difference value from the plurality of calculated difference values, the one having the smallest difference value is finally determined as the true difference value. As a result, even when there is an error such that the current data becomes distance data of any point within a predetermined range in the vertical and horizontal directions with the coordinate (x, y) as the center, the background due to the error The influence of the blur of the upper position can be minimized.

  As described above, according to the seventh embodiment, the current data calculation unit 20 that acquires the current data from the distance data acquired by the three-dimensional laser scanner 10 and the coordinates of the distance data acquired by the three-dimensional laser scanner 10. Or the comparison data calculating part 30 which acquires the comparison data located in the predetermined range set centering on the pixel, the current data which the current data calculating part 20 acquired, and the several comparison which the comparison data calculating part 30 acquired Since the true difference value is calculated from the difference value with the data, and the first change area extraction unit 40a that extracts the change area from the calculated true difference value is provided, the position where the laser light pulse reaches Even in an environment that includes an error, accurate distance information between the object and the background can be acquired. Thereby, the recognition accuracy of the object can be improved.

  Further, according to the seventh embodiment, when the point at which the laser light pulse arrives is shifted within a range of ± 1 point in the vertical and horizontal directions around the coordinates of the current data due to an error of the dispersion mechanism 13, for example. Since the difference value is calculated using the nine comparison data around the current data and the true difference value is selected from the difference value, the deviation of the point where the laser light pulse reaches It is possible to select a stable difference value that suppresses the influence of. Thereby, even when the position where the laser light pulse reaches due to an error in distance data is blurred, the influence of the blur can be suppressed to the minimum. Therefore, the recognition accuracy of the object can be improved.

  In the seventh embodiment described above, as an example, the configuration in which the comparison data is acquired from the coordinates located in the range of ± 1 point in the up / down / left / right diagonal direction with the current data as the center has been described. The error occurring at the upper position is not limited to “± 1 point”, but varies. Therefore, the range of the comparison data to be acquired varies depending on the error that occurs at the position on the background where the laser light pulse reaches. Moreover, although the structure which acquires comparison data on the basis of a coordinate was shown, it is good also as a structure which acquires comparison data on the basis of a pixel.

  In the above-described seventh embodiment, the case where the first change region extraction unit 40a is applied to the configuration illustrated in the first embodiment has been described. However, the configuration of the seventh embodiment is illustrated in the second embodiment. It is also applicable to other configurations. In that case, the first change region extraction unit 40 shown in FIG. 6 is replaced with the first change region extraction unit 40a described above. Further, in the flowchart of FIG. 7, the process of step ST5 is replaced with the process of step ST91 to step ST94 described above. The processing after step ST31 is performed on the true difference value acquired at step ST94.

Embodiment 8 FIG.
In the above-described seventh embodiment, a configuration has been described in which the smallest difference value among the difference values between the current data and each comparison data is acquired as a true difference value. Here, acquiring the minimum difference value as a true difference value means that the sensitivity of extracting the change region is low. If the subject who has entered the field of view is the notification target, the sensitivity degradation due to the small difference value may make the discovery of the notification target insensitive, and the risk of misreporting for the notification target Occurs. In view of this, the eighth embodiment shows a configuration that improves the extraction sensitivity of the change region when a notification target appears in the field of view.

FIG. 29 is a block diagram showing the configuration of the monitoring apparatus according to the eighth embodiment.
The monitoring device 100k according to the eighth embodiment additionally includes a determination result accumulation unit 40b in the first change area extraction unit 40a of the monitoring device 100j according to the seventh embodiment shown in FIG. In the following description, the same or corresponding parts as those of the monitoring apparatus 100j according to the seventh embodiment are denoted by the same reference numerals as those used in the first embodiment, and the description thereof is omitted or simplified.

  As in the seventh embodiment, the comparison data calculation unit 30 sets the distance data to be accumulated around the coordinates or pixels of the distance data input from the three-dimensional laser scanner 10. The distance data of the points within the range is converted into comparison data and stored in the comparison data storage unit 31.

  The determination result accumulation unit 40b is an area for storing past determination results of the first change area extraction unit 40a. The determination result of the first change area extraction unit 40a is a determination result of whether or not the true difference value is greater than or equal to a threshold value. Either a certain coordinate is extracted as a change area or a certain coordinate is not a change area. It is information indicating whether or not it has been determined. The determination result accumulation unit 40b accumulates the determination result at each coordinate or each pixel at least one frame before.

  The first change area extraction unit 40a acquires the current data accumulated in the current data accumulation unit 21 and all comparison data accumulated in the comparison data accumulation unit 31, and coordinates the current data and all comparison data. Alternatively, a plurality of difference values are calculated by comparison in units of pixels. The first change region extraction unit 40a refers to the determination result storage unit 40b and determines whether the true difference value at the same coordinate one frame before is determined to be equal to or greater than the threshold and is extracted as a change region.

  When it is determined that the true difference value is equal to or greater than the threshold value, the first change area extraction unit 40a acquires the maximum difference value among the difference values calculated at the coordinates as the true difference value. On the other hand, when it is determined that the true difference value is not greater than or equal to the threshold value, the first change region extraction unit 40a acquires the smallest difference value among the difference values calculated at the coordinates as the true difference value. . The first change area extraction unit 40a extracts, as a change area, a coordinate area or a pixel area whose acquired true difference value is greater than or equal to a preset threshold value.

Next, the operation of the monitoring device 100k according to the eighth embodiment will be described.
FIG. 30 is a flowchart showing the operation of the monitoring apparatus according to the eighth embodiment.
In the following, the same steps as those of the monitoring device 100j according to the seventh embodiment are denoted by the same reference numerals as those used in FIG. 28, and the description thereof is omitted or simplified. Further, as in the seventh embodiment, the case where the error occurring at the position on the background is “± 1 point” will be described as an example.

  When the difference value for each coordinate is calculated in step ST93, the first change area extraction unit 40a refers to the determination result accumulation unit 40b, and the coordinate for which the difference value is calculated in step ST93 changes in the image one frame before. It is determined whether or not the area is determined (step ST101). When it is determined that the region is a change region in the image one frame before (step ST101; YES), the first change region extraction unit 40a sets the maximum difference value among the difference values for each coordinate calculated in step ST93 to be true. (Step ST102). On the other hand, when it is determined that it is not a change area in the image one frame before (step ST101; NO), the first change area extraction unit 40a calculates the minimum difference value among the difference values for each coordinate calculated in step ST93. A true difference value is set (step ST102).

  Since the true difference value obtained in step ST101 or step ST102 is 8-bit multi-value data, the first change area extraction unit 40a determines whether or not the obtained true difference value is equal to or greater than a preset threshold value. A determination is made (step ST95). If the true difference value is greater than or equal to the threshold (step ST95; YES), the pixel area is extracted as a change area (step ST7). On the other hand, when the true difference value is less than the threshold value (step ST95; NO), it is determined that the pixel area is not a change area (step ST8), and the process proceeds to step ST9.

With the above-described configuration, when a certain coordinate or pixel is extracted as a change area, the maximum difference value is always selected as the true difference value, and thus the true difference value is always the maximum value. This means that the sensitivity for extracting the change region is high. On the other hand, when a certain coordinate or pixel is not extracted as the change area, since no object exists at the coordinate, the minimum difference value is selected as the true difference value. This means that the true difference value is always the minimum value, and the sensitivity for extracting the change region is low.
By these processes, the maximum difference value can be selected when it is extracted as a change area, and if the notification target appears in the field of view, the change area extraction sensitivity can be improved. , To suppress the misreporting for the notification target.

  As described above, according to the eighth embodiment, the current data calculation unit 20 that acquires current data from the distance data acquired by the three-dimensional laser scanner 10 and the coordinates of the distance data acquired by the three-dimensional laser scanner 10. Alternatively, it is a determination result of the comparison data calculation unit 30 that acquires comparison data located within a predetermined range set around the pixel and the first change region extraction unit 40a for the image one frame before, A determination result storage unit 40b that stores a determination result as to whether or not each coordinate or each pixel of the image before the frame is a change area, and the current data and comparison data positioned within a predetermined range centering on the current data A difference value is calculated, and it is determined whether or not the coordinates for which the difference value has been calculated with reference to the determination result storage unit 40b is a change area in the image one frame before, and according to the determination result Since the first change area extraction unit 40a for determining the true difference value is provided, the maximum difference value is acquired as the true difference value for the coordinate or pixel that was the change area in the image one frame before. It is possible to improve the extraction sensitivity of the change area. Thereby, the detection capability of a report object can be improved and the risk of unreporting can be reduced. Moreover, the recognition accuracy of the object can be improved.

  In the above-described eighth embodiment, as an example, the configuration in which the comparison data is acquired from the coordinates located in the range of ± 1 point in the vertical, horizontal, and diagonal directions around the current data has been described. The error occurring at the upper position is not limited to “± 1 point”, but varies. Therefore, the range of the comparison data to be acquired varies depending on the error that occurs at the position on the background where the laser light pulse reaches. Moreover, although the structure which acquires comparison data on the basis of a coordinate was shown, it is good also as a structure which acquires comparison data on the basis of a pixel.

  In the above-described eighth embodiment, the case where the first changed region extraction unit 40a and the determination result storage unit 40b are applied to the configuration illustrated in the first embodiment has been described. However, the configuration of the eighth embodiment is as follows. The present invention can also be applied to the configuration shown in Embodiment Mode 2. In that case, the first change region extraction unit 40 shown in FIG. 6 is replaced with the first change region extraction unit 40a and the determination result storage unit 40b described above. Further, in the flowchart of FIG. 7, the process of step ST5 is replaced with the process of step ST91 to step ST103 described above. The processing after step ST31 is performed on the true difference value acquired at step ST102 or step ST103.

  Each function of the monitoring apparatus described in the first to eighth embodiments is realized by a processing circuit. The processing circuit can realize the functions described above by hardware, software, firmware, or a combination thereof. Software and firmware are described as programs and stored in a memory. The processing circuit executes the function of each unit by reading and executing the program stored in the memory.

  Since the monitoring apparatus according to the present invention can recognize an object with high accuracy, the monitoring apparatus is suitable for applying to a monitoring camera or the like to accurately recognize a notification target and issue a warning.

  10 Three-dimensional laser scanner, 11 Laser light emitting unit, 12 Laser light pulse, 13 Dispersion mechanism, 14 Dispersed laser light pulse, 15 Laser reflected light, 16 Laser light receiving unit, 20 Current data calculation unit, 21 Current data storage unit, 30 Comparison Data calculation unit, 31 Comparison data storage unit, 40, 40a First change region extraction unit, 40b Determination result storage unit, 41 Positive / negative map matching unit, 42 Second change region extraction unit, 43 Stability confirmation unit, 43a storage Unit, 50 recognition processing unit, 60 notification processing unit, 70, 70a distance calculation unit, 71, 71a distance histogram creation unit, 72 object distance peak point calculation unit, 73 background distance peak point calculation unit, 74, 74a distance difference calculation Part, 75 distance difference connection processing part, 80 intensity calculation part, 81 intensity histogram creation part, 82 object strength Peak point calculation unit, 83 Background intensity peak point calculation unit, 84 Intensity difference calculation unit, 85 Intensity difference connection processing unit, 86 Third change region extraction unit, 100, 100a, 100b, 100c, 100d, 100e, 100f, 100g , 100h, 100i monitoring device, 200 background, 201 object.

Claims (9)

From the measurement result of the three-dimensional laser scanner that measured the monitoring area, information on the distance to the monitoring area is acquired, and the current data calculation unit is used as the current distance data;
A comparison data calculation unit that obtains distance information from the measurement result to the monitoring region and converts it into comparison distance data;
A difference value between the current distance data acquired by the current data calculation unit and the comparison distance data converted by the comparison data calculation unit is calculated, and a region where the difference value is equal to or greater than a threshold is extracted as a change region. A change area extraction unit ,
The first change area extraction unit includes map data storing a difference value to be rejected based on a distance relationship between the background and the object in the monitoring area, and a difference value between the current distance data and the comparison distance data. And a positive / negative map matching unit that corrects a difference value rejected by the map data . A distance information from the measurement result to the monitoring area is acquired, a first distance histogram indicating a short-time change of the monitoring area, and a long-time change of the monitoring area that is a time change longer than the first distance histogram A distance histogram creation unit for creating a second distance histogram indicating
An object distance peak point calculation unit for calculating an object distance peak point from the first distance histogram;
A background distance peak point calculation unit for calculating a background distance peak point from the second distance histogram;
A distance difference calculation unit that calculates a difference peak point by subtracting the background distance peak point from the object distance peak point;
The monitoring apparatus according to claim 1, further comprising: a second change area extraction unit that extracts a change area from an area based on the distance information of the difference peak point calculated by the distance difference calculation unit. Intensity information indicating the reflectance of the monitoring area is acquired from the measurement result, and a first intensity histogram indicating a short-time change of the monitoring area and a time change of the monitoring area that is longer than the first intensity histogram. An intensity histogram creation unit for creating a second intensity histogram showing a long-term change;
An object intensity peak point calculation unit for calculating an object intensity peak point from the first intensity histogram;
A background intensity peak point calculation unit for calculating a background intensity peak point from the second intensity histogram;
An intensity difference calculation unit for calculating a difference peak point obtained by subtracting the background intensity peak point from the object intensity peak point;
An intensity difference connection processing unit that connects the intensity information of the difference peak points calculated by the intensity difference calculation unit;
The monitoring apparatus according to claim 1, further comprising: a third change area extraction unit that extracts the change area from the intensity information connected by the intensity difference connection processing unit. A distance information to the monitoring area is acquired from a measurement result of the three-dimensional laser scanner that measured the monitoring area, and a first distance histogram indicating a short time change of the monitoring area and a time longer than the first distance histogram A distance histogram creating unit that creates a second distance histogram indicating a long-term change in the monitoring area that is a change;
An object distance peak point calculation unit for calculating an object distance peak point from the first distance histogram;
A background distance peak point calculation unit for calculating a background distance peak point from the second distance histogram;
A distance difference calculation unit that calculates a difference peak point by subtracting the background distance peak point from the object distance peak point;
The monitoring apparatus provided with the 2nd change area extraction part which extracts a change area from the area | region based on the distance information of the difference peak point which the said distance difference calculation part calculated. From the measurement result of the three-dimensional laser scanner that measured the monitoring area, the intensity information indicating the reflectance of the monitoring area is acquired, and the first intensity histogram indicating the short-time change of the monitoring area and the first intensity histogram An intensity histogram creating unit for creating a second intensity histogram indicating a long-term change in the monitoring area, which is a longer time change;
An object intensity peak point calculation unit for calculating an object intensity peak point from the first intensity histogram;
A background intensity peak point calculation unit for calculating a background intensity peak point from the second intensity histogram;
An intensity difference calculation unit for calculating a difference peak point obtained by subtracting the background intensity peak point from the object intensity peak point;
An intensity difference connection processing unit that connects the intensity information of the difference peak points calculated by the intensity difference calculation unit;
The monitoring apparatus provided with the 3rd change area extraction part which extracts a change area from the intensity | strength information which the said intensity | strength difference connection process part connected. A distance histogram creation unit that obtains distance information to the monitoring region from a measurement result of the three-dimensional laser scanner that measures the monitoring region, and creates a first distance histogram indicating a short-time change of the monitoring region;
An object distance peak point calculation unit for calculating an object distance peak point from the first distance histogram;
A background distance peak information storage unit that stores a distance peak point of an object that is fixedly present in the monitoring area;
A distance difference calculation unit for calculating a difference peak point obtained by subtracting the distance peak point accumulated in the background distance peak information accumulation unit from the object distance peak point;
The monitoring apparatus provided with the 2nd change area extraction part which extracts a change area from the area | region based on the distance information of the difference peak point which the said distance difference calculation part calculated. An accumulation unit for accumulating the change region extracted by the third change region extraction unit;
4. The apparatus according to claim 3, further comprising a stability confirmation unit that determines that the change area is a person when the change amount of the change area accumulated in the accumulation unit before and after the elapse of a predetermined time is greater than or equal to a threshold value. Or the monitoring apparatus of Claim 5 . The comparison data calculation unit acquires distance information from a position within a range set around the measurement result to the monitoring area and converts it into the comparison distance data,
The first change area extraction unit calculates a difference value between the current distance data acquired by the current data calculation unit and the comparison distance data converted by the comparison data calculation unit, and the smallest of the calculated difference values The monitoring apparatus according to claim 1, wherein the difference value is a true difference value, and an area where the true difference value is equal to or greater than a threshold is extracted as a change area. The comparison data calculation unit acquires distance information from a position within a range set around the measurement result to the monitoring area and converts it into the comparison distance data,
The first change area extraction unit calculates a difference value between the current distance data acquired by the current data calculation unit and the comparison distance data converted by the comparison data calculation unit, and the past change area extraction results Based on the above, it is determined whether the maximum difference value among the calculated difference values is a true difference value or the minimum difference value is a true difference value, and the determined true difference value is equal to or greater than a threshold value The monitoring apparatus according to claim 1, wherein an area that is a change area is extracted.

Images