JP5187878B2 - Object measurement system - Google Patents

Description

  The present invention relates to an object measurement system, and more particularly to a measurement system for a moving object such as a pedestrian moving within a predetermined area, a method for processing object information, and a program for the processing.

  When examining the sales situation of a product and the consumer's product preference, it is important to grasp the movement of a person in a specific place, for example, a supermarket. In addition, it may be necessary to investigate the movement of objects such as people and vehicles in an amusement park or event plaza.

  Conventionally, it has been proposed to use a laser range finder when detecting an object or measuring the surface of an object. For example, as a technique for measuring the surface shape of an object, Patent Document 1 discloses a technique in which a focus light source device is attached to a stereo camera and the object shape is calculated by image calculation. Patent Document 2 discloses a technique for measuring the surface shape of an object by incorporating a mechanism for scanning a laser beam in a three-dimensional direction inside the laser distance measuring device and performing a three-dimensional scan. Further, in Patent Document 3, two laser distance measuring devices are installed at a high position in a measurement region, and the two laser irradiation surfaces are parallel to irradiate the surface substantially perpendicularly to the ground. A technique for measuring the speed and shape of a vehicle passing through the vehicle is disclosed. Furthermore, Patent Document 4 discloses a technique for measuring the distance around an object with an arm in which laser displacement meters are installed at different heights.

JP 2001-194128 A JP 2002-181533 A JP 2001-319290 A JP 7-260440 A

  However, the technique of Patent Document 1 has a problem that it is necessary to perform complicated image calculation, and the cost required for calculating the three-dimensional position of the object is high. In addition, a special device called a stereo camera with a focal light source is required, and it is often difficult to apply in a real environment where the influence of ambient light is strong. In addition, since the viewing angle is narrow, it is not suitable for a wide range of measurements.

  The technique disclosed in Patent Document 2 incorporates a mechanism for three-dimensionally scanning a laser inside the laser distance measuring device, which complicates the internal structure and has a problem with durability. Further, since the laser irradiation direction is operated in a complicated manner, it takes time for one scan, and the object to be measured needs to be stationary for a certain time until the scan is completed in the measurement region. Therefore, it is difficult to measure a moving object at high speed.

  The technique of the above-mentioned patent document 3 mainly targets a relatively flat object such as a vehicle, and when irradiating a vertically elongated object such as a person in the vertical direction, mainly the upper surface of the object. That is, there is a problem in that only information on the person's head can be obtained. The object speed is calculated only at the moment of entering and exiting the laser measurement section, but the speed changes while passing through the measurement section, or the object entry direction is orthogonal to the irradiation surface. When there is no direction, it is difficult to calculate an accurate object shape.

  The technique of the above-mentioned Patent Document 4 is expensive because the technique of measuring the periphery of an object with an arm provided with a laser displacement meter requires the use of a large-scale device. Further, since the measurement needs to be performed inside the rotating arm, there is a limitation on the size of the object that can be measured depending on the size of the arm.

Therefore, an object of the present invention is to acquire information on an object moving within a measurement region using a laser distance measuring device without cost and labor for measurement, and to obtain the position and three-dimensional shape data of the object. It is easy to get.
Further, the present invention is to display a three-dimensional shape of a specified object using the acquired object information.

An object measurement system according to the present invention is an object measurement system for measuring an object moving in an area using a laser distance measuring device,
A laser distance measuring device that irradiates the area to be measured with laser and obtains distance measurement data related to position information about an object in the area;
A processing device that performs data processing on the distance measurement data obtained from the laser distance measuring device, and storage means for storing related information;
The processing device is configured to obtain, from the ranging data continuously obtained from the laser ranging device, three-dimensional coordinates in a common coordinate system defined with the laser ranging device as an origin for each point data of an object by laser irradiation. Seeking data ,
Further, for each point data of the object obtained from this point, the points that are present within a certain distance from the three-dimensional coordinate data in the common coordinate system and are regarded as the same object.
A first processing means for performing processing to make a cluster;
The cluster is mapped to one while obtaining the centroid coordinates of a plurality of points included in one cluster after the cluster processing is performed, and the identification information specific to the object is given to the mapped cluster by obtaining the centroid coordinates. Second processing means to:
3 in the local coordinate system defined by using the center of gravity coordinates of the object obtained by the second processing means as the origin from the three-dimensional coordinate data of the object obtained by the first processing means in the common coordinate system. Third processing means for obtaining dimensional coordinate data;
The storage means has an object position DB for storing the barycentric coordinate data of the object obtained by the third processing means in correspondence with the identification information of the object. Composed.

Object information processing method according to the present invention is an object information processing method obtained by measuring the moving objects laser range equipment using area,
Irradiating the area to be measured with a laser to obtain distance measurement data relating to position information about an object in the area;
From the distance measurement data obtained continuously from the laser distance measurement device, three-dimensional coordinate data in a common coordinate system defined with the laser distance measurement device as the origin for each point data of the object by laser irradiation ,
Furthermore, for each point data of the object obtained from this, a first process for clustering points that are within a certain distance from the three-dimensional coordinate data in the common coordinate system and are regarded as the same object within the laser point group. Processing steps;
The cluster is mapped to one while obtaining the centroid coordinates of a plurality of points included in one cluster after the cluster processing is performed, and the identification information specific to the object is given to the mapped cluster by obtaining the centroid coordinates. A second processing step,
From the three-dimensional coordinate data of the object obtained by the first processing step in the common coordinate system, 3 in the local coordinate system defined by using the barycentric coordinates of the object obtained by the second processing step as the origin. A third processing step for obtaining dimensional coordinate data;
And storing the barycentric coordinate data of the object obtained in the third processing step in an object position DB corresponding to the identification information of the object. The

The program according to the present invention is a program executed on a computer for processing information of an object obtained by measuring an object moving within an area using a laser distance measuring device,
Laser irradiation is performed on the region to be measured from the laser distance measuring device, and the laser distance measuring device is defined with the laser distance measuring device as the origin for each point data of the object by laser irradiation based on distance measurement data relating to position information obtained continuously. Find 3D coordinate data in a common coordinate system ,
Further, a first process for clustering points that are present within a certain distance from the three-dimensional coordinate data in the common coordinate system and are regarded as the same object for each point data of the object obtained from this point data. Steps,
The cluster is mapped to one while obtaining the centroid coordinates of a plurality of points included in one cluster after the cluster processing is performed, and the identification information specific to the object is given to the mapped cluster by obtaining the centroid coordinates. A second processing step,
From the three-dimensional coordinate data of the object obtained by the first processing step in the common coordinate system, 3 in the local coordinate system defined by using the barycentric coordinates of the object obtained by the second processing step as the origin. A third processing step for obtaining dimensional coordinate data;
A program for executing, on a computer, the step of storing the barycentric coordinate data of the object obtained in the third processing step in the object position DB corresponding to the identification information of the object Is done.

According to the present invention, it is possible to obtain distance measurement data by irradiating a measurement area with a laser using a laser distance measuring apparatus, and to acquire the position and three-dimensional shape data of an object moving within the measurement area. In addition, by irradiating the target with the laser at an angle, three-dimensional shape data of various heights of the object can be acquired as the object moves. Further, the three-dimensional shape of the object can be displayed using the acquired object information.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 shows a configuration of a moving object measuring system according to an embodiment.
This system determines a specific area such as a supermarket or an amusement park entrance as the measurement area X, and targets a pedestrian as a moving object to be measured. This is a system that measures the shape, size, speed, etc. of a pedestrian who has moved within the measurement region X, and stores the information in a DB in order to use it for, for example, market research. Thereby, it becomes possible to acquire information such as how many pedestrians are accompanied, whether they are children or adults, whether they have luggage, whether they are checking carts, or the like.

  As shown in the figure, this system receives two laser distance measuring devices 11 and 12 that irradiate the measurement region X, and distance data detected by the laser distance measuring devices 11 and 12, and moves including a pedestrian. It is composed of one PC (personal computer) that calculates the three-dimensional shape of an object. The PC 3 is a CPU 31 as a processing device for calculating and storing the surface shape, position and moving speed of a moving object, a database (DB) 32 for storing various related data, a display device 33 and an input device 34. Have

The laser distance measuring device 1 (hereinafter, when the two devices 11 and 12 are collectively referred to as “1”) is a device that continuously measures a single plane with a rotating laser. For example, the laser distance measuring device 11 is installed at a height of about 16 cm from the floor surface, and the laser distance measuring device 12 is obliquely downward at a depression angle θ (pitch angle P _B ) from a height of about 2 m. Installed with the irradiation surface tilted.
The laser distance measuring device 1 continuously measures the distance to the object while changing the angle in the range of 180 ° in front of the fan in steps of 0.5 °, and information including the distance data to the moving object and the angle number thereof. Send to PC3. The distance data is configured as a set of data in which an angle obtained by engraving a range of 180 ° by 0.5 ° and a distance to the front of the angle are arranged in centimeters, as in Expression (1).

  The data for one scan is a data string of distance values obtained by one rotation of the laser, and distance data of several tens of frames per second is transmitted to the PC 3.

The measurement range defines a common coordinate system W (x _w , y _w , z _w ) with the laser distance measuring device 11 as the origin (xyz coordinates (0, 0, 0)), and the laser distance measuring device in that coordinate system 11 position (x _A , y _A , z _A ), posture (roll angle R _A , pitch angle P _A , azimuth angle Y _A ) and position of laser ranging device 12 (x _B , y _B , z _B ), Postures (roll angle R _B , pitch angle P _B , azimuth angle Y _B ) are obtained in advance. Here, the roll angle is a rotation angle with respect to an axis perpendicular to the laser front surface, the pitch angle is a rotation angle with respect to an axis perpendicular to the laser right side, and an azimuth angle is rotation with respect to an axis perpendicular to the laser right above. It is a horn. Further, the measurement range of the laser distance measuring device 12 is installed so as to be included in the measurement range of the laser distance measuring device 11. A range measured by the laser distance measuring devices 11 and 12 is referred to as a measurement region, and a pedestrian entering the region is set as a measurement target.

  FIG. 2 shows an example of data obtained by one scan of the laser distance measuring devices 11 and 12 when a pedestrian 20 enters the measurement area. (A) and (B) show data obtained from the laser distance measuring devices 11 and 12, respectively.

Laser light emitted from the laser distance measuring device 11 always strikes the vicinity of the pedestrian's ankle. Therefore, as shown in (A), the arc-shaped laser point data is obtained from the front surface of both feet of the pedestrian 20, and the position of the pedestrian in the measurement region is known.
Moreover, as shown in (B), the height at which the laser light from the laser distance measuring device 12 irradiates the pedestrian changes according to the distance between the laser distance measuring device 12 and the pedestrian. When the pedestrian is far away from the laser distance measuring device 12, the laser light hits the lower position of the pedestrian, and the laser light hits the higher position of the pedestrian as the pedestrian approaches the laser distance measuring device 12.

As shown in FIG. 3A, as the pedestrian 20 moves from the right to the left in the measurement region, the laser beam from the laser distance measuring device 12 is irradiated from the foot to the head, and the surface shape of the pedestrian is related. A plurality of three-dimensional coordinate data can be obtained. For example, when the scanning timing of the laser beam is t1 to t6, as shown in (B), the surface shape data C1 to C6 can be obtained as a plurality of three-dimensional coordinates for the pedestrian. When a large number of three-dimensional coordinate data collected in this way is processed and displayed on the display device 33, the entire surface shape of a pedestrian can be displayed as shown in FIG. This system acquires and displays the surface shape of the pedestrian based on such a principle. Note that FIG. 3 shows six surface shape data C1 to C6 for the sake of explanation, but if the number of laser light scans is increased, three-dimensional display to the extent that the pedestrian (moving object) can be fully recognized is possible. is there.

A specific configuration of the DB formed in the DB 32 will be described with reference to FIGS.
FIG. 8 shows the configuration of the three-dimensional coordinate management DB 800.
The three-dimensional coordinate management DB 800 manages the measured three-dimensional coordinate data for each object in order to grasp the three-dimensional shape of the object (pedestrian). In order to identify a pedestrian who has entered the measurement area, an object ID is assigned to each pedestrian, and a point of the object and a three-dimensional coordinate corresponding to the point are stored for each object ID. Point No is a serial number assigned to each point.
For example, 801 stores the three-dimensional coordinates of the object ID: 00001, and 802 and 803 store the three-dimensional coordinates of the object IDs: 00002 and 00003, respectively.

FIG. 9 shows the configuration of the object ID management DB 900.
The object DI management DB 900 is a DB for managing the correspondence between the movement time of each pedestrian and the shape data. The object DI management DB 900 corresponds to the object ID, the entry time to the measurement region X, the exit time, and the object. The address of the three-dimensional coordinate data is stored. With this address, the three-dimensional coordinate data corresponding to the object ID can be read from the three-dimensional coordinate management DB 800.

FIG. 10 shows the configuration of the object position DB 100.
The object position DB 100 manages the movement of pedestrians and stores barycentric coordinates g1 corresponding to the object ID. When recognizing the same pedestrian, the center-of-gravity coordinate g1 is updated each time. On the other hand, when recognizing a new pedestrian, a new object ID is assigned to the pedestrian. The barycentric coordinates g1 are stored.

A series of processing of the object measurement system according to the present invention is realized by executing a program by the CPU 31 of the PC 3.
The overall processing in this system will be described. As shown in FIG. 14, first, background data of a measurement location is acquired (S1401), and then surface shape data (three-dimensional shape) of a moving object entering the measurement location. Data) is acquired (S1402). When the surface shape data of the moving object is accumulated, display processing of the moving object is subsequently performed (S1403). The surface shape data acquisition process will be described later with reference to FIG. 4 and the display process will be described later with reference to FIG.

Here, background data acquisition processing (S1401) will be described. When this system is started, first, background data of a measurement place is acquired as a process before measuring a pedestrian. For example, the entire measurement place where there is no pedestrian is irradiated from the laser distance measuring device 1 to acquire data (background data) related to an object that does not move, such as a wall or a pillar as a background. When the distance data acquired by the laser distance measuring device 1 is sequentially stored in the DB 32 of the PC 3 and a histogram of distance values is calculated for each angle of the distance data, the laser point that has continued to hit the non-moving object takes the mode value. Therefore, the mode value becomes background data (see FIG. 5B).
The acquired background data is stored in the DB 32 in a format in which angles and distances to the background are arranged in order, similarly to the distance data shown in Expression (1).

Next, with reference to FIG. 4, a process for acquiring three-dimensional shape data will be described.
The following processing is performed by executing a program with the CPU 31 of the PC 3.
First, the object ID is set to “0” (S0). The object ID is information for identifying that the detected object is the same, and a new object ID is given every time it is recognized that a new pedestrian has entered the measurement region.
Next, the PC 3 stands by until it becomes possible to receive the distance data of each of the laser distance measuring devices 11 and 12 (S1), and when it can be received, it receives two distance data (S2). The received distance data is compared with the background data, respectively, and only the data in which the moving object such as a pedestrian hits the laser beam is calculated (S3).
That is, as shown in FIG. 5, the distance data (A) detected by the laser distance measuring device 1 includes background data 51 such as walls 53 and pillars in the measurement region and pedestrian data 52. When the CPU 31 of the PC 3 calculates the difference of the background data 51 from the received distance data, the CPU 31 calculates the moving object data Cn (FIG. 3) in which the background data 51 is deleted and only the pedestrian data remains. When there is a pedestrian in the measurement area, one or more data of the moving object data Cn remain, and when there is no pedestrian in the measurement area, the distance data and the background data are equal and no background difference remains. Become.

Next, the moving object data C obtained in (S1) is converted from the polar coordinate system of the angle and distance to the planar coordinate system (X, Y), and further the position (x _A , y _{A of the} laser distance measuring device 1). , Z _A ) and posture (R _A , P _A , Y _A ) are converted into three-dimensional coordinates (x, y, z) in the common coordinate system W (S4). The conversion formula (Formula 2) is shown below.

Here, alpha respective angles indicated by the background difference, r is the distance indicated by angle, the conversion formula, similarly to the laser range finder 12, the position of the laser distance measuring device 11 (x _A, y _a, z _a), the attitude _{(R a, P a, Y} a) position of the laser distance measuring device 12 _{(x B, y B, z} B), attitude (R _B, P _B, replaced by Y _B) Use the same thing. Hereinafter, the coordinate data groups obtained by converting the moving object data C for one scan into coordinates (x, y, z) in the common coordinate system W are referred to as position data and shape data, respectively.
At this time, it is determined whether there is a pedestrian based on the presence or absence of shape data (S5). As a result of the determination, if there is no shape data (S5: No), the CPU 31 waits for the next distance data to be sent again.

  On the other hand, for example, when there is a moving object in the shape data obtained from the laser distance measuring device 12, the CPU 31 performs a clustering process of position data (S6). Clustering is a process in which laser points existing within a certain distance in a laser point group are grouped as one object (cluster). By performing this recursively, the position data based on the two feet of the same pedestrian are combined into one as the same person, and distinguished from another pedestrian that is more than a certain distance.

After clustering, the barycentric coordinates g1 of all points included in one cluster are obtained (S7). Here, one coordinate is determined as the center of gravity of the cluster, and is set as a position where a pedestrian is present.
After obtaining the center-of-gravity coordinates of all the clusters, the object position DB 100 is searched for the center of gravity of each cluster, so that the object position DB 900 for each cluster of the center-of-gravity coordinates g1 that is less than a certain distance (for example, 15 cm or less). To obtain a related object ID (S8).

As a result, when there is an associated object ID and it is determined that they are the same person (S9), a movement vector is obtained from the elapsed time from the previous scan and the movement distance of the center of gravity (S10), and the corresponding object position DB 100 corresponds. The center-of-gravity coordinates g1 obtained by calculating the center of gravity of the object ID in step S7 is updated. (S11)
Similarly for the shape data, clustering is performed with the centroid coordinates g1 obtained in step S7 as the starting point, and the surface shape of the object measured by the laser distance measuring device 1 of the pedestrian detected from the position data is obtained ( S12).

  As shown in FIG. 7 (B ′), a local coordinate system L is defined with the center of gravity coordinate g1 obtained in step S7 as the origin, and the coordinate group associated with the same person is obtained in step S11 and obtained in step S11. The direction of the movement vector v is converted into coordinates rotated in the positive direction of the y-axis (S13). This conversion is calculated using the following equation (3).

Before conversion of the common coordinate system at the indicated coordinates in the shape data in step _{S11 (x W, y W,} z W), the barycentric coordinates of the clusters obtained in step S7 (xg, yg, zg) , in step S10 When the angle formed between the x _W axis in a common coordinate system W and the movement vector and phi, the coordinates projected to the local coordinate system L (x ', y', z ') is obtained from equation (3).

The converted coordinates are added and registered in the three-dimensional coordinate management DB 800 for each object ID as three-dimensional coordinate data of the object shape (S14).
On the other hand, if a cluster determined to be the same person is not found in step S9 (S9: No), it is determined that a new pedestrian has entered, and a new object ID is given to the pedestrian. (Adding “1” to the latest object ID) is created (S91), and the barycentric coordinates g1 obtained in step S7 are newly added to the object position DB 100 and registered (S92). At this time, the current time is registered in the new object ID and entry time fields in the object ID management DB 900 of FIG. 9 (S93).

  The object position DB 100 is searched for the center-of-gravity coordinates g1 of all the clusters, and the association of the object ID is searched (S15). As a result, the cluster of the object ID that is not associated with any cluster is obtained from the object position DB 100. The object ID is discarded. Also, the current time is registered in the exit time column of the object ID management DB 900, and an address for accessing the 3D coordinate data of the 3D coordinate management DB 800 is registered in the address column of the object ID management DB 900 (S16).

  As described above, by executing the processing from step 1 to step 16, the three-dimensional shape for each pedestrian arriving in the measurement area X can be recorded in the three-dimensional coordinate management DB 900, and the series of processing is completed. To do.

Here, the process of step S6 to step S13 will be described more specifically.
As shown in FIG. 6, when a moving object exists in the laser data obtained from the laser distance measuring device 11 irradiated horizontally (that is, when there is difference data in (C) background difference in FIG. 5), the background of the data The laser points after obtaining the difference are clustered (S6), and the barycentric coordinates of the points included in the cluster are obtained. Here, the center-of-gravity coordinate of the cluster at time t1 is g1, and the coordinate is the position of the pedestrian at time t1.

  Next, clustering is similarly performed at time t2 to obtain the center-of-gravity coordinates g2 of the cluster, and pedestrians existing within a certain range from the position g1 of the pedestrian at time t1 are determined as the same person (S9), t1 , t2 is obtained from the difference in barycentric coordinates between times t2 (S10).

  Next, as shown in FIG. 7, in the data obtained from the laser distance measuring device 12 at time t2, the laser laser points existing within the threshold from the centroid coordinates g2 of the pedestrian calculated by the above processing are clustered (S12). ). Further, a local coordinate system L is defined with the center of gravity coordinate g2 as the origin (0, 0, 0), the coordinate vector clustered in step S12 is rotated in the positive direction of the y-axis, and the local coordinate system L1 is rotated. Projection is performed on the system L for conversion (S13).

Also at the next time t3, the local coordinate system L centered on the center-of-gravity coordinate g3 is updated by similarly obtaining the center-of-gravity coordinate g3 and the movement vector v2 of the object from the laser distance measuring device 11. Similarly to the above, the laser data obtained from the laser distance measuring device 12 at time t3 is projected and converted to the local coordinate system L using Expression (3).
As described above, the surface shape data of the pedestrian projected onto the local coordinate system L is sequentially stored in the three-dimensional coordinate management DB 800 for each object ID.

Next, an object shape display process (S4003) in the present system will be described with reference to FIG. The following processing is performed by executing a program with the CPU 31 of the PC 3.
After the measurement of the moving object is completed, the input device 34 of the PC 11 is operated to display the initial screen for displaying the object shape on the display device 33 (S1501). As shown in FIG. 16 (FIG. 16 is a screen showing a search result), this screen includes an input field 1601 for “search section”. The operator inputs a time range to be searched for in this input field 1601 and presses the search button.

  When the CPU 31 determines that a search operation has been performed by pressing the search button (S1502), the CPU 31 accepts data in the search section. When the data of the search section is accepted, the CPU 31 determines whether the data is correct (S1503). This determination is to check whether the format of the input value is correct. If the determination result is correct, the object ID management DB 900 is searched to obtain the object ID, the entry time, the exit time, and the address of the three-dimensional coordinate data of the object that has entered the measurement area in the designated search section (S1504). . Then, the object ID of the search result and the entry time and exit time of the object ID into the measurement area are displayed on the screen of the display device 33 (S1505).

As shown in FIG. 16, for example, as a search section 1601, a result of searching the object ID management DB 900 (FIG. 9) in which a section between 2006/04/12, 13:02:00 to 13:05:00 is designated. Three object IDs (00003, 00004, 00005) are hit, and the search result 1602 is displayed.
Here, if no object has entered the measurement area in the search section and the object ID does not hit, the process returns to the initial screen and waits for the input of the “search section” and the pressing of the search button. (S1506).

On the other hand, when one or more object IDs are hit, the operator moves the cursor 1603 from the search result 1602 and selects a target object ID. When the CPU 31 detects that an object ID ('00005') has been selected (S1507), it accesses the 3D coordinate management DB 800 with an address associated with the object ID and reads the corresponding 3D coordinate data ( S1508). FIG. 17 shows an example of the read three-dimensional coordinate data.
When reading the three-dimensional coordinate data, the CPU 31 creates a three-dimensional scatter diagram based on the data and displays a screen showing the shape of the object (person) (S1509). FIG. 17 shows a display example.

The process of creating a screen showing the shape of the object using the three-dimensional coordinate data can be realized by using an existing program (S1509).
FIG. 18 shows a display example of the object shape. The screen includes an object information display 1801 such as an object ID of the displayed object, an entry time to the measurement region, and an exit time, and a three-dimensional object display 1802. On the screen, there are prepared a “re-search button” 1803 for stopping the display of the object, returning to the initial screen and redoing the search, and an “end button” 1804 for terminating the system.

  Here, the object shape display processing (steps S1509 to 1513) can be displayed with various modifications. This object shape display is three-dimensionally processed as shown in FIG. 18, but when the user operates the input device 34, the three-dimensional figure is appropriately rotated (S 1510, S 1511) and enlarged. Or can be reduced (S1512, S1513), and the display viewpoint can be changed. For example, FIG. 19 is a display example when the display of FIG. 18 is “rotated 90 degrees left”, and the shape of the left side of the object can be displayed at this viewpoint.

  Thus, the shape of the object can be analyzed by searching and displaying the object that entered the measurement area at a certain time, and rotating, enlarging, or reducing the display shape. Further, the display of the object is stopped by operating the “re-search button” 1803, the search can be performed again by returning to the initial screen (1514), or the system can be ended by operating the “end button” 1804. Yes (1515).

Next, another example will be described with reference to FIGS.
FIG. 12 shows a configuration in which the laser distance measuring device 12 in FIG. 1 is installed on the ground and the measurement surface is directed obliquely upward. By installing in this way, it is not necessary to install the laser distance measuring device 12 at a high place, and installation work and maintenance become easy.
Further, as shown in FIG. 13, the three-dimensional information and positions of many objects can be obtained by arranging the laser distance measuring device 13 at another position and measuring the distance using three or more laser distance measuring devices 13. Information can be obtained. By adding the laser distance measuring device 13, when the pedestrian moves to the front with respect to the laser distance measuring device 12, not only the front surface of the pedestrian but also the shape of the rear surface of the pedestrian is obtained by the laser distance measuring device 13. be able to.

As mentioned above, although the preferable Example of this invention was described, this invention is not limited to an above-described Example, It can implement in various deformation | transformation.
For example, in the above example, a three-dimensional shape of a person is acquired. However, according to another example, the moving object to be measured is not limited to a person, and distance measurement for an arbitrary object such as a vehicle or an animal is possible. Is possible. Further, even when there are a plurality of objects in the measurement region, it is possible to measure the three-dimensional shape of each object at the same time.

Moreover, the structure of DB shown in FIGS. 8-10 is an example, and is not limited to this. Further, the terminology used in the above embodiments is an example and is not limited to the above expression. For example, the PC may be a computer or a calculation means in short. The DB may be referred to as a table, a storage area, or storage means in terms of expression. In the above embodiment, for the sake of explanation, the three DBs 800 to 100 are provided. However, as long as they are related by the object ID, these may be combined into two, or may be combined into four or more DBs. You may make it comprise and divide | segment.
In addition, the moving object display process is not limited to the process performed by the PC 3, and may be performed by another server or PC. In this case, of course, the related DBs 800 and 900 can be accessed from these devices.
Further, this display process may be performed at any time. For example, the event may be performed at night after the event of the day ends.

  As described above, according to the present embodiment, the target is irradiated with the laser at an angle, so that the laser beam irradiated at an angle is irradiated at various heights on the side surface of the object as the object moves. For this reason, it is not necessary to move the laser irradiation direction in a complicated manner as in the three-dimensional scanning method, and the operation part inside the apparatus is simple and hardly broken. Therefore, the time required for one scan is short, and the three-dimensional shape of the object can be acquired at high speed.

Furthermore, by irradiating the laser obliquely, the shape of the whole body can be acquired for an object that is elongated in the vertical direction like a person. In addition, by calculating the position and speed of the object by horizontal irradiation and appropriately transforming the shape of the side surface of the object obtained by oblique irradiation based on the information, accurate three-dimensional independent of speed and moving direction. Get the shape.
Since this system can be measured by at least two laser distance measuring devices, the system can be constructed in a small scale, simply and inexpensively. In addition, it is possible to flexibly cope with the object size simply by changing the laser irradiation angle or the installation position.

  Also, information such as the width and height of the object can be obtained from the information more easily and at a higher speed than the three-dimensional shape of the object. Thus, an application application in which such information is added to the shape can be considered. For example, the age and sex of a person can be estimated from the shape and speed.

The figure which shows the structural example of the moving object measurement system in one Example, A diagram showing an overview of data obtained from a pedestrian moving in the measurement area, The figure which shows notionally the mode of the three-dimensional shape data obtained in a time series from a laser range finder when a pedestrian moves a measurement area, The flowchart figure which shows the processing operation for acquiring the three-dimensional shape data in a moving object measurement system, Conceptual diagram showing the relationship between distance data and background data in the measurement region, The figure which shows the process which converts into the local coordinate system from the data obtained from the laser distance measuring device 11. The figure which shows the process which converts into the local coordinate system from the data obtained from the laser distance measuring device 12. The figure which shows the storage format of three-dimensional coordinate management DB800, The figure which shows the storage format of object ID management DB900, The figure which shows the storage format of object position DB100, The conceptual diagram which shows the mode of the laser ranging device 11 and the step of a pedestrian in the measurement area | region, The figure which shows the modification of arrangement | positioning of the laser ranging apparatus in FIG. The figure which shows the modification of arrangement | positioning of the laser ranging apparatus in FIG. The flowchart figure which shows the whole processing operation in a moving object measurement system, The flowchart figure which shows the display processing operation | movement of the moving object in a moving object measurement system, The figure which shows the example of a display of the search result of object ID management DB900, The figure which shows the data structural example in 3D coordinate management DB800, The figure which shows the example of a display of the object shape processed three-dimensionally, The figure which shows the example of a display at the time of rotating a display object 90 degree | times left.

Explanation of symbols

X: Measurement area 1, 11, 12: Laser distance measuring device 3: PC 31: CPU 32: DB 33: Display device 34: Input device 800: Three-dimensional coordinate management DB 900: Object ID management DB 100: Object position DB

Claims (6)

An object measurement system for measuring an object moving in an area using a laser range finder,
A laser range finder that irradiates the area to be measured with laser and obtains distance measurement data related to the distance from the object in the area;
A processing device for performing data processing on the ranging data obtained from the laser ranging device;
Having storage means for storing relevant information;
The storage means stores an object position DB that stores barycentric coordinate data of the object corresponding to identification information unique to the object;
In order to grasp the three-dimensional shape of the object, it has a three-dimensional coordinate management DB for storing the obtained three-dimensional coordinate data of the object corresponding to the identification information of the object,
The processing device has common coordinates defined with the laser distance measuring device as an origin for each point data indicating the position and shape of the object by laser irradiation from the distance measurement data continuously obtained from the laser distance measuring device. Find 3D coordinate data in the system,
Further, from the data indicating the position of the object in the three-dimensional coordinate data in the common coordinate system for each point data of the object obtained from this point, points that are present within a certain distance and regarded as the same object in the laser point group First processing means for performing processing to form a cluster;
After obtaining the barycentric coordinates of a plurality of points included in one cluster after performing the cluster processing and mapping the cluster to one as a position of the object, and obtaining the barycentric coordinates of all the clusters The same identification information as that of the object is given to the cluster whose centroid coordinates are within a certain distance, and new identification information is given to the object for the cluster whose centroid coordinates are not within a certain distance. Two processing means;
From the three-dimensional coordinate data in the common coordinate system of the object determined by the first processing means, in the local coordinate system defined with the barycentric coordinates of the object determined by the second processing means as the origin, Third processing means for obtaining three-dimensional coordinate data indicating the shape data of the object by converting the moving direction of the object into coordinates rotated around a predetermined three-dimensional axis;
The object position DB updates the barycentric coordinates for the same identification information with respect to the barycentric coordinates obtained by the second processing means, and the obtained barycentric coordinates for the newly given identification information. Corresponding to the new identification information,
The three-dimensional coordinate management DB stores the three-dimensional coordinate data of the object obtained by the third processing unit corresponding to the identification information of the object,
The object position DB is searched to obtain the barycentric coordinate data corresponding to the specified identification information of the object, and the three-dimensional coordinate management DB is further searched to correspond to the identification information of the object. An object measurement system characterized in that three-dimensional coordinate data is obtained and an object is measured by grasping shape data of the object indicated by the three-dimensional coordinate data at a position indicated by the barycentric coordinate data. The storage means accesses the time when the object is measured and the barycentric coordinate data of the object stored in the object position DB, corresponding to the identification information of the object given by the second processing means. And an object ID management DB for storing addresses capable of
The object measurement system according to claim 1 . The first processing means calculates the three-dimensional coordinate data of the object in the common coordinate system from the distance measurement data using the equation (2),
The object measurement system according to claim 1 or 2, wherein the third processing means calculates the barycentric coordinate data of the object using the equation (3).

Formula (2)

Formula (3)

A laser distance measuring device that obtains distance measurement data related to a distance from an object in the region by irradiating the region to be measured, and a processing device that performs data processing on the distance measurement data obtained from the laser distance measuring device And an object information processing method in an object measurement system having storage means for storing related information,
Irradiating the area to be measured with a laser to obtain distance measurement data relating to position information about an object in the area;
Three-dimensional coordinates in a common coordinate system defined with the laser distance measuring device as the origin for each point data indicating the position and shape of the object by laser irradiation from the distance measurement data continuously obtained from the laser distance measuring device Seeking data,
Further, from the data indicating the position of the object in the three-dimensional coordinate data in the common coordinate system for each point data of the object obtained from this point, points that are present within a certain distance within the laser point group and are regarded as the same object are obtained. A first processing step for performing a clustering process;
After obtaining the barycentric coordinates of a plurality of points included in one cluster after performing the cluster processing and mapping the cluster to one as a position of the object, and obtaining the barycentric coordinates of all the clusters The same identification information as that of the object is given to the cluster whose centroid coordinates are within a certain distance, and new identification information is given to the object for the cluster whose centroid coordinates are not within a certain distance. Two processing steps;
From the three-dimensional coordinate data in the common coordinate system of the object obtained by the first processing step, in a local coordinate system defined using the barycentric coordinates of the object obtained by the second processing step as an origin, A third processing step of obtaining three-dimensional coordinate data indicating shape data of the object by converting the moving direction of the object into coordinates rotated around a predetermined three-dimensional axis;
With respect to the barycentric coordinates obtained by the second processing step, the barycentric coordinates are updated for the same identification information of the object position DB, and the obtained barycentric coordinates are used for the newly assigned identification information. Storing in the object position DB corresponding to new identification information;
Storing the three-dimensional coordinate data of the object obtained by the third processing step in a three-dimensional coordinate management DB corresponding to the identification information of the object;
The object position DB is searched to obtain the barycentric coordinate data corresponding to the specified identification information of the object, and the three-dimensional coordinate management DB is further searched to correspond to the identification information of the object. Obtaining three-dimensional coordinate data, grasping shape data of the object indicated by the three-dimensional coordinate data at a position indicated by the obtained barycentric coordinate data, and measuring the object. Object information processing method. Furthermore, the identification information of the object is designated by operating the input device, the barycentric coordinate data corresponding to the identification information of the designated object is obtained, and the three-dimensional coordinate management DB is further searched to obtain the object The three-dimensional coordinate data corresponding to the identification information is obtained, and the display device performs drawing processing using the shape data of the object indicated by the three-dimensional coordinate data at the position indicated by the obtained barycentric coordinate data The object information processing method according to claim 4, wherein the object information processing method is displayed. A program executed on a computer for processing information on an object obtained by measuring an object moving within an area using a laser distance measuring device,
Irradiating the area to be measured with a laser to obtain distance measurement data relating to position information about an object in the area;
Three-dimensional coordinates in a common coordinate system defined with the laser distance measuring device as the origin for each point data indicating the position and shape of the object by laser irradiation from the distance measurement data continuously obtained from the laser distance measuring device Seeking data,
Further, from the data indicating the position of the object in the three-dimensional coordinate data in the common coordinate system for each point data of the object obtained from this point, points that are present within a certain distance within the laser point group and are regarded as the same object are obtained. A first processing step for performing a clustering process;
After obtaining the barycentric coordinates of a plurality of points included in one cluster after performing the cluster processing and mapping the cluster to one as a position of the object, and obtaining the barycentric coordinates of all the clusters The same identification information as that of the object is given to the cluster whose centroid coordinates are within a certain distance, and new identification information is given to the object for the cluster whose centroid coordinates are not within a certain distance. Two processing steps;
From the three-dimensional coordinate data in the common coordinate system of the object obtained by the first processing step, in a local coordinate system defined using the barycentric coordinates of the object obtained by the second processing step as an origin, A third processing step of obtaining three-dimensional coordinate data indicating shape data of the object by converting the moving direction of the object into coordinates rotated around a predetermined three-dimensional axis;
With respect to the barycentric coordinates obtained by the second processing step, the barycentric coordinates are updated for the same identification information of the object position DB, and the obtained barycentric coordinates are used for the newly assigned identification information. Storing in the object position DB corresponding to new identification information;
Storing the three-dimensional coordinate data of the object obtained by the third processing step in a three-dimensional coordinate management DB corresponding to the identification information of the object;
The object position DB is searched to obtain the barycentric coordinate data corresponding to the specified identification information of the object, and the three-dimensional coordinate management DB is further searched to correspond to the identification information of the object. Obtaining three-dimensional coordinate data, grasping the shape data of the object indicated by the three-dimensional coordinate data at the position indicated by the obtained barycentric coordinate data, and measuring the object on the computer A program characterized by

Images