JP7493154B2 - Information processing device and counting method - Google Patents

Description

This disclosure relates to an information processing device and a counting method.

In recent years, people flow (number of people) surveys have become important in marketing research and traffic volume surveys. Conventionally, there are tally counters that count people by hand. In addition, Patent Document 1 discloses a method of counting people using a camera.

Japanese Patent Application Laid-Open No. 62-34293

Tally counters are inefficient because they rely on human hands. Patent Document 1 uses a camera to count people, so there are cases where the number of people cannot be counted properly due to changes in brightness between day and night, for example.

Non-limiting examples of the present disclosure contribute to providing an information processing device and counting method that can easily and appropriately count the number of objects that pass through a measurement boundary line set on a passageway using radio waves.

An information processing device according to an embodiment of the present disclosure includes a grouping unit that groups point cloud data measured by a radar device that emits radio waves toward a passageway and associates the number of objects included in the group with the group, and a passage determination unit that counts the number of objects associated with the group that have passed through a measurement boundary line set on the passageway.

A counting method according to one embodiment of the present disclosure groups point cloud data measured by a radar device that emits radio waves toward a passageway, associates the number of objects contained in the group with the group, and counts the number of objects associated with the group that pass through a measurement boundary line set on the passageway.

These comprehensive or specific aspects may be realized as a system, device, method, integrated circuit, computer program, or recording medium, or as any combination of a system, device, method, integrated circuit, computer program, and recording medium.

According to one embodiment of the present disclosure, the number of objects that pass through a measurement boundary line set on a passageway can be counted easily and appropriately using radio waves.

Further advantages and benefits of an embodiment of the present disclosure will become apparent from the specification and drawings. Such advantages and/or benefits may be provided by some of the embodiments and features described in the specification and drawings, respectively, but not necessarily all of them need be provided to obtain one or more identical features.

FIG. 1 is a diagram illustrating an example of a counting system using an information processing device according to the present disclosure. A diagram explaining an example of human tracking operations FIG. 13 is a diagram illustrating an example of grouping processing. FIG. 13 is a diagram for explaining an example of group separation processing. FIG. 13 is a diagram for explaining an example of group merging processing; FIG. 1 is a diagram for explaining an example of processing by direction. A diagram showing an example of the block configuration of an information processing device and a server. A flowchart showing an example of processing of an information processing device. A flowchart showing an example of processing of an information processing device.

Below, the embodiments of the present disclosure will be described in detail with appropriate reference to the drawings. However, more detailed explanation than necessary may be omitted. For example, detailed explanations of already well-known matters and duplicate explanations of substantially identical configurations may be omitted. This is to avoid the following explanation becoming unnecessarily redundant and to make it easier for those skilled in the art to understand.

The accompanying drawings and the following description are provided to enable those skilled in the art to fully understand the present disclosure, and are not intended to limit the subject matter described in the claims.

FIG. 1 is a diagram showing an example of a counting system using an information processing device 2 according to the present disclosure. As shown in FIG. 1, the counting system includes a radar device 1, an information processing device 2, and a server 3.

In FIG. 1, the measurement boundary line A1 is shown by a dashed line. The measurement boundary line A1 is set, for example, on a passage such as a road where the traffic volume of people and vehicles is measured. The measurement boundary line A1 is set, for example, in a program executed by the information processing device 2. The position of the measurement boundary line A1 may be changed by the user.

The information processing device 2 counts people and vehicles passing through the measurement boundary line A1 by distinguishing the direction in which they pass through the measurement boundary line A1. For example, the information processing device 2 counts people and vehicles passing through the measurement boundary line A1 in the direction of the arrow A2a in FIG. 1 (the direction moving toward the radar device 1). The information processing device 2 counts people and vehicles passing through the measurement boundary line A1 in the direction of the arrow A2b in FIG. 1 (the direction moving away from the radar device 1).

The radar device 1 is installed on a structure A3, such as a utility pole. The radar device 1 emits radio waves toward the passageway and generates point cloud data based on the waves reflected from objects such as people and vehicles. The radar device 1 transmits the generated point cloud data to the information processing device 2, for example, by wireless communication.

The information processing device 2 identifies people and vehicles (targets) based on the point cloud data transmitted from the radar device 1. The information processing device 2 counts the number of identified people and vehicles that pass through the measurement boundary line A1. As described above, the information processing device 2 distinguishes the direction in which the people and vehicles pass through the measurement boundary line A1 and counts the number of people and vehicles that pass through the measurement boundary line A1. The information processing device 2 transmits the counted number of people and vehicles to the server 3, for example, via the wired network 4.

The server 3 analyzes the number of people and vehicles transmitted from the information processing device 2. For example, the server 3 analyzes the cumulative number of people and vehicles that pass through the measurement boundary line A1 per unit time (e.g., per hour or per day). The server 3 displays the analysis results on a display device, for example, as shown in the example screen A4 in FIG. 1.

In this way, the information processing device 2 counts the number of people and vehicles using the point cloud data of the radar device 1. This allows the information processing device 2 to count the number of people and vehicles without relying on human hands, as with a tally counter. In addition, because the point cloud data is acquired using radio waves, the information processing device 2 can appropriately count people and vehicles regardless of brightness.

In the above, the radar device 1 transmits the point cloud data to the information processing device 2 using wireless communication, but this is not limited to this. The radar device 1 may transmit the point cloud data to the information processing device 2 using wired communication.

In addition, the information processing device 2 transmits the counted numbers of people and vehicles to the server 3 via the wired network 4, but this is not limited to the above. The information processing device 2 may transmit the counted numbers of people and vehicles to the server 3 using wireless communication.

The information processing device 2 may be incorporated into the radar device 1 or into the server 3. For example, the information processing device 2 may have its functions incorporated into the radar device 1 as a program or a semiconductor chip. The information processing device 2 may have its functions incorporated into the server 3 as a program.

In addition, multiple measurement boundaries A1 may be set along the passage.

We will explain how to count people. Compared to vehicles, people tend to move in groups (close together). For example, this applies when several people are walking and talking.

When tracking each person in a crowded area and counting the number of people passing through the measurement boundary line A1, the processing load on the information processing device becomes large.

Figure 2 is a diagram explaining an example of the operation of tracking a person. In Figure 2, the measurement boundary line A1 is shown by a solid line. The numbers 1 to 3 and questions shown in Figure 2 represent people. The black dots represent point cloud data of people measured by the radar device 1.

In Figure 2, people move from top to bottom. Time passes in the order of (A), (B), and (C) in Figure 2. In other words, Figure 2 shows the movement of people along an aisle in the order of (A), (B), and (C). The dotted circle in Figure 2 indicates the position of the person at the previous processing (time).

As shown in Figure 2 (A), three people are crowded together and move toward the measurement boundary line A1.

As shown in FIG. 2B, three people are positioned close together on the measurement boundary line A1. When three people are close together, the radar device 1 cannot separate them due to resolution issues.

When people cannot be separated due to the resolution of the radar device 1, the information processing device may not be able to properly track individual people, for example, as shown in the question in FIG. 2(B). For example, the point cloud data shown in the question in FIG. 2(B) actually shows two people, but the information processing device may determine that there is only one person. In this way, if the information processing device cannot properly track individual people, it will not be able to properly count the number of people passing through the measurement boundary line A1.

As shown in FIG. 2C, three people crowded together pass through measurement boundary line A1 and then move away. The information processing device may not be able to properly track point cloud data in which individual people cannot be properly tracked (the people in the two questions in FIG. 2B). For example, the point cloud data shown by the questions in solid circles in FIG. 2C actually shows two people, but the information processing device may determine that there is only one person.

In this way, when people are moving in a crowded area, the information processing device may not be able to properly count the number of people passing through the measurement boundary line A1.

Therefore, the information processing device 2 in FIG. 1 groups people who are crowded together and counts the number of people in each group who pass through the measurement boundary line A1.

Figure 3 is a diagram illustrating an example of grouping processing. Figure 3 (A), (B), and (C) show the movement of people in a time series, similar to Figure 2. The detection records shown in Figure 3 (A), (B), and (C) show information about people or groups in each time series of Figure 3 (A), (B), and (C). The detection records are stored, for example, in a memory unit of an information processing device. The detection records may be considered as information linked to a group.

As shown in FIG. 3A, when people are crowded together, the information processing device 2 groups the crowded people. In the example of FIG. 3A, the information processing device 2 groups three people into one group, as shown in the dashed frame A11. The information processing device 2 may group parts of the point cloud data where the density is high, for example, using DBSCAN (Density-Based Spatial Clustering of Applications with Noise).

When the information processing device 2 groups people who are crowded together, it assigns an ID (identifier) to each group. In the example of FIG. 3A, the information processing device 2 assigns ID1, as shown in the detection record of FIG. 3A.

In addition, the information processing device 2 stores the number of people in the group of ID1, the previous (previous) location of the group of ID1, and the current location of the group of ID1. In the example of FIG. 3A, the information processing device 2 stores three people, the previous location x=0, y=12, and the current location x=0, y=11, as shown in the detection record of FIG. 3A.

The position of the group may be the position of the center of gravity of the group. For example, the position of the group may be the position of the center of gravity of the area indicated by the dashed frame A11 in FIG. 3A. The previous position indicates the position of the group in the previous detection process, and the current position indicates the position of the group in the current detection process (the detection process in FIG. 3A).

The information processing device 2 stores the change between the previous relative position of the group with respect to the measurement boundary line A1 and the current relative position of the group with respect to the measurement boundary line A1. In the previous detection process, the relative position of the group ID1 with respect to the measurement boundary line A1 is set to "up". In the current detection process (the detection process in (A) of Figure 3), the relative position of the group ID1 with respect to the measurement boundary line A1 is "up", so the information processing device 2 stores "up" from "up" as shown in the detection record in (A) of Figure 3.

In the example of Figure 3, if the group is located on the radar device 1 side of the measurement boundary line A1, the group's relative position with respect to the measurement boundary line A1 is defined as "top". If the group is located on the measurement boundary line A1, the group's relative position with respect to the measurement boundary line A1 is defined as "center". If the group is located in the area opposite the area on the side where the radar device 1 is located with respect to the measurement boundary line A1, the group's relative position with respect to the measurement boundary line A1 is defined as "bottom". Unless otherwise specified below, the group's relative position with respect to the measurement boundary line A1 is defined as described above.

In addition, when the information processing device 2 determines (detects) a group for the first time in the detection process of FIG. 3(A), the previous position does not exist. Therefore, the previous position is not stored in the detection record. In addition, when the information processing device 2 determines a group for the first time in the detection process of FIG. 3(A), the previous relative position does not exist. Therefore, the only change in relative position is the current relative position.

As shown in FIG. 3B, the group ID1 (the center of gravity of the group) is located directly above the measurement boundary line A1. The detection record at this time is as shown in FIG. 3B.

The previous position shown in the detection record in Figure 3(B) indicates the position measured in the previous detection process (previous detection process) of the group. Therefore, the previous position shown in the detection record in Figure 3(B) matches the current position shown in the detection record in Figure 3(A).

The relative position change shown in the detection record in Figure 3(B) has gone from "top" to "center." The "top" of the relative position change shown in the detection record in Figure 3(B) matches the "top" on the right side of the relative position change shown in the detection record in Figure 3(A).

As shown in FIG. 3C, the group with ID1 is located below the measurement boundary line A1. The detection record at this time is as shown in FIG. 3C.

The previous position shown in the detection record in Figure 3(C) indicates the position measured immediately before the group. Therefore, the previous position shown in the detection record in Figure 3(C) matches the current position shown in the detection record in Figure 3(B).

The relative position change shown in the detection record in Figure 3(C) is from the "center" to "down." The "center" of the relative position change shown in the detection record in Figure 3(C) coincides with the "center" of the relative position change shown in the detection record in Figure 3(B).

In this way, when people are crowded together, the information processing device 2 groups the crowded people. The information processing device 2 tracks moving people in groups, not individually. Then, when a group passes through the measurement boundary line A1, the information processing device 2 counts the number of people who pass through the measurement boundary line A1 based on the number of people included in the group.

For example, in the example of FIG. 3, when the group of ID1 passes through the measurement boundary line A1, the information processing device 2 counts that three people have passed through the measurement boundary line A1 based on the number of people included in the information (detection record) of the group of ID1. This allows the information processing device 2 to appropriately count the number of people passing through the measurement boundary line A1.

The information processing device 2 determines whether the group has passed the measurement boundary line A1 based on the relative position change included in the detection record. In other words, the information processing device 2 determines whether the group has passed the measurement boundary line A1 based on the positional relationship of the current position and the previous position with respect to the measurement boundary line A1.

For example, the information processing device 2 determines that the group has passed the measurement boundary line A1 when the relative position change indicates "center" to "bottom", "center" to "top", "top" to "bottom", or "bottom" to "top". In other words, when the relative position change in the detection record corresponds to any of the four patterns described above, the information processing device 2 counts the number of people included in the detection record as the number of people who have passed the measurement boundary line A1. In the example of Figure 3, when (C) in Figure 3 is shown, the information processing device 2 determines that the group of ID1 has passed the measurement boundary line A1, and counts that three people have passed the measurement boundary line A1.

Groups can split up. For example, when a group of people walking and talking decides to leave the group.

Figure 4 is a diagram illustrating an example of group separation processing. Similar to Figure 3, Figure 4 (A), (B), and (C) show the movement of people in a time series. Similar to Figure 3, the detection records shown in Figure 4 (A), (B), and (C) show information about people or groups in each time series of Figure 4 (A), (B), and (C).

As shown in Figure 4 (A), the group with ID1 is located on the measurement boundary line A1. The number of people in the group with ID1 is three, as shown in the detection record in Figure 4 (A). The previous position is x = 0, y = 12. The current position is x = 0, y = 11. The relative position change is from "above" to "above."

As shown in the dashed frame A21 in FIG. 4B, when people are crowded together, the information processing device 2 groups the crowded people. In the example of FIG. 4B, the information processing device 2 groups two people into one group, as shown in the dashed frame A21.

When the information processing device 2 groups people who are crowded together, it assigns an ID to the group. In the example of FIG. 4B, the information processing device 2 assigns ID2 to the group shown in the dashed frame A21, as shown in the detection record of FIG. 4B.

In addition, the information processing device 2 also assigns an ID to a person determined to be a single person (the group of point cloud data shown in the dashed frame A22) as shown in the dashed frame A22 in FIG. 4B. In the example of FIG. 4B, the information processing device 2 assigns ID1 to the person shown in the dashed frame A22 as shown in the detection record in FIG. 4B.

Here, if the person or group to which the ID has been assigned is located within a predetermined distance and is also located within a predetermined distance from the group detected in the previous detection process, the information processing device 2 determines that the person or group to which the ID has been assigned has separated from one group.

For example, in the example of FIG. 4B, the person with ID1 and the group with ID2 are located within a predetermined distance. Also, the person with ID1 and the group with ID2 are located within a predetermined distance from the group detected in the previous detection process (1 in the dotted circle in FIG. 4B). In this case, the information processing device 2 determines that the person with ID1 and the group with ID2 belonged to one group (belonging to the group with ID1 shown in the dashed frame in FIG. 4A) in the previous detection process (the detection process in FIG. 4A) and have been separated from that group.

When the information processing device 2 determines that a person or group has separated from one group, it regroups the people or groups as shown in the solid-line frame A23 in (B) of Figure 4, and transfers part of the information from the previous detection record to the information from the current detection record of the regrouped group.

For example, the information processing device 2 transfers the current position of the detection record in (A) of FIG. 4 to the position immediately before the detection record in ID1 in (B) of FIG. 4, and transfers the current position of the detection record in (A) of FIG. 4 to the position immediately before the detection record in ID2 in (B) of FIG. 4. In addition, the information processing device 2 transfers one of the three people in ID1 in (A) of FIG. 4 to the number of people in ID1 in (B) of FIG. 4, and transfers two of the three people in ID1 in (A) of FIG. 4 to the number of people in ID2 in (B) of FIG. 4.

As shown in Figure 4 (B), the person with ID1 is located directly above the measurement boundary line A1. The group with ID2 (center of gravity) is located below the measurement boundary line A1. Therefore, the relative position change of ID1 shown in the detection record of Figure 4 (B) is from "top" to "center." The relative position change of ID2 is from "top" to "bottom." Note that the "top" of the relative position change shown in the detection record of ID1 and ID2 in Figure 4 (B) matches the "top" on the right side of the relative position change shown in the detection record of Figure 4 (A).

The relative position change from "top" to "bottom" is determined to have passed the measurement boundary line A1, as explained in FIG. 3. Therefore, the information processing device 2 determines that the group with ID2, whose relative position change is from "top" to "bottom," has passed the measurement boundary line A1, and counts it as 2.

As shown in FIG. 4C, the person with ID1 is located below the measurement boundary line A1. The group with ID2 is located below the measurement boundary line A1. Therefore, the relative position change of ID1 shown in the detection record in FIG. 4C is from "center" to "bottom." The relative position change of ID2 is from "bottom" to "bottom."

The "center" of the relative position change shown in the detection record of ID1 in Figure 4(C) coincides with the "center" of the relative position change shown in the detection record of ID1 in Figure 4(B). The "bottom" of the relative position change shown in the detection record of ID2 in Figure 4(C) coincides with the "bottom" of the relative position change shown in the detection record of ID2 in Figure 4(B).

The relative position change from "center" to "down" is determined to have passed the measurement boundary line A1, as explained in FIG. 3. Therefore, the information processing device 2 determines that the group of ID1, whose relative position change is from "center" to "down," has passed the measurement boundary line A1, and counts it by 1.

In this way, when the information processing device 2 determines that a person or group has separated from one group, the detection information of the original group is passed on to the person or group after separation. This allows the information processing device 2 to appropriately count the number of people passing through the measurement boundary line A1.

For example, the separated person and group (person with ID1 and group with ID2) shown in FIG. 4B inherit the current position of the group they were separated from in FIG. 4A as their position immediately before the detection record. This allows the information processing device 2 to properly determine that the group shown in dashed frame A21 in FIG. 4B has passed through the measurement boundary line A1, and that the person shown in dashed frame A22 has not passed through the measurement boundary line A1, and thus allows the number of people passing through the measurement boundary line A1 to be properly counted.

In the example of FIG. 4, one group is separated into one person and one group, but this is not limited to this. One person may be separated into multiple groups from one group. Multiple groups may be separated from one group.

One group may combine with another. For example, two separate groups may be walking and talking and then join together.

Figure 5 is a diagram illustrating an example of group merging processing. (A), (B), (C), and (D) of Figure 5 show the movement of people in a time series, similar to Figure 3. The detection records shown in (A), (B), (C), and (D) of Figure 5 show information about people or groups in each time series of (A), (B), (C), and (D) of Figure 5, similar to Figure 3.

As shown in the dashed frame A31 and the detection record in FIG. 5(A), the information processing device 2 assigns ID1 to a person who is moving alone. The previous position of the person with ID1 is x=0, y=14. The current position of the person with ID1 is x=0, y=13. The relative position change of the person with ID1 is from "above" to "above."

Furthermore, as shown in the dashed frame A32 and the detection record in FIG. 5A, the information processing device 2 groups two people and assigns them ID2. The previous position of the group with ID2 is x=0, y=13. The current position of the group with ID2 is x=0, y=12. The relative position change of the group with ID2 is from "above" to "above".

As shown in the dashed frame A31 in Figure 5(B), the person with ID1 moves to just before the measurement boundary line A1. The position immediately before the person with ID1 matches the current position of the detection record of ID1 in Figure 5(A), and is x = 0, y = 13. The current position of the person with ID1 is x = 0, y = 11. The relative position change of the person with ID1 is from "above" to "above".

As shown in dashed frame A32 in Figure 5(B), the ID2 group moves below the measurement boundary line A1. The previous position of the ID2 group coincides with the current position of the detection record of ID2 in Figure 5(A), and is x = 0, y = 12. The current position of the ID2 group is x = 0, y = 9. The relative position change of the ID2 group is from "up" to "down."

The person with ID1 shown in Figure 5(B) moves under the measurement boundary line A1 as shown in Figure 5(C). The group with ID2 shown in Figure 5(B) moves further under the measurement boundary line A1 as shown in Figure 5(C).

In FIG. 5C, a person with ID1 is combined with a group with ID2. When a person or group is combined, the information processing device 2 assigns one ID to the combined group. In the example of FIG. 5C, the information processing device 2 assigns ID1, as shown in the detection record of FIG. 5C.

If the group assigned ID1 is located within a predetermined distance from multiple people or groups detected in the previous detection process, the information processing device 2 determines that the group assigned ID1 is a combined group.

For example, in the example of Figure 5 (C), the group with ID1 (1 in the solid circle in Figure 5 (C)) is located within a predetermined distance from the person detected in the previous detection process (1 in the dotted circle at the bottom of Figure 5 (C)). Also, the group with ID1 (1 in the solid circle in Figure 5 (C)) is located within a predetermined distance from the group detected in the previous detection process (2 in the dotted circle at the bottom of Figure 5 (C)). In this case, the information processing device 2 determines that the group shown in the dashed frame A33 in Figure 5 (C) is a combined group.

When the information processing device 2 determines that a group merge has occurred, it inherits the detection information in the detection record before the merger (the previous detection record).

For example, the number of people 1 shown in the detection record of Figure 5 (C) inherits the number of people in the detection record of ID 1 in Figure 5 (B). The previous position of the number of people 1 shown in the detection record of Figure 5 (C) inherits the current position of ID 1 in Figure 5 (B).

The number of people 2 shown in the detection record of Figure 5 (C) inherits the number of people in the detection record of ID2 in Figure 5 (B). The previous position of the number of people 2 shown in the detection record of Figure 5 (C) inherits the current position of ID2 in Figure 5 (B).

The current position of the detection record of the combined group indicates the position of the combined group. In the example of FIG. 5C, the position of the combined group shown in the dashed frame A33 is x=0, y=8, as shown in the detection record of FIG. 5C.

The relative position change for number of people 1 shown in the detection record in Figure 5(C) is from "top" to "bottom." The "top" relative position change for number of people 1 shown in the detection record in Figure 5(C) inherits the "top" on the right side of the relative position change for ID1 in Figure 5(B).

The relative position change for number of people 2 shown in the detection record in Figure 5(C) is from "down" to "down." The "down" on the left side of the relative position change for number of people 2 shown in the detection record in Figure 5(C) inherits the "down" on the right side of the relative position change for ID2 in Figure 5(B).

As shown in FIG. 5D, group 1 of ID1, which was combined in the detection process in FIG. 5C, moves further down relative to measurement boundary line A1. When the relative position changes contained in the detection records become common, the information processing device 2 consolidates the number of people and position information in the detection records.

For example, the information processing device 2 makes the relative position change of the group with ID1 shown in FIG. 5(D) common from "below" to "below," as shown in the detection record in FIG. 5(D). The information processing device 2 sets the number of people in the detection record to "3," which is the total number of people in the immediately preceding detection process, as shown in the detection record in FIG. 5(D). The information processing device 2 sets the current position of the immediately preceding detection process, x=0, y=8, to a common immediately preceding position, as shown in the detection record in FIG. 5(D). The information processing device 2 makes the current position of the group with ID1 common to x=0, y=7, as shown in the detection record in FIG. 5(D).

In this way, when people or groups are merged, the information processing device 2 inherits the detection information from the detection records of each person or group before they were merged. This allows the information processing device 2 to appropriately count the number of people passing through the measurement boundary line A1.

For example, in FIG. 5B, the information processing device 2 counts the number of people in the group shown in dashed frame A32 based on the relative position change in the detection record of ID2 from "above" to "below". In the merged group shown in FIG. 5C, the information processing device 2 counts only the people corresponding to dashed frame A31 before the merger based on the relative position change in number of people 1 in the detection record from "above" to "below". In the merged group in FIG. 5C, the information processing device 2 does not count the number of people in number of people 2 of the group corresponding to dashed frame A32 before the merger. This allows the information processing device 2 to appropriately count the number of people passing through the measurement boundary line A1.

In the example of FIG. 5, one person and a group are combined, but this is not limited to this. Multiple people may be combined, or multiple groups may be combined.

When groups moving in different directions are crowded together, the position of the group may become unclear. For example, one group moves from above the measurement boundary line to below, and another group moves from below the measurement boundary line to above. In this case, when one group is crowded together with another group, the position of the group becomes unclear.

Figure 6 is a diagram explaining an example of processing by direction. In Figure 6, (A), (B), and (C) of Figure 6 show the movement of people from left to right in time series.

The dashed frame A41 shown in FIG. 6A indicates a group of two people. The dashed frame A42 shown in FIG. 6A indicates a group of three people. The group in dashed frame A41 moves from above to below the measurement boundary line A1, and the group in dashed frame A42 moves from below to above the measurement boundary line A1.

As shown in the center of the time series in FIG. 6A, when the group shown in the dashed frame A41 and the group shown in the dashed frame A42 are crowded together, the information processing device 2 combines these two groups. The two groups are different groups that move in different directions, and the information processing device 2 may not be able to properly determine the movement of the center of gravity of the combined group.

The information processing device 2 therefore executes the counting process of the number of people passing through the measurement boundary line A1, dividing it into two layers based on the direction of movement of the people or groups. For example, the information processing device 2 separates it into a first layer that counts people or groups moving from above to below the measurement boundary line A1, and a second layer that counts people or groups moving from above to below the measurement boundary line A1. The information processing device 2 executes the counting process of the number of people passing through the measurement boundary line A1 in each of the two layers.

(B) in FIG. 6 shows an example of processing in the first layer. In the first layer, the information processing device 2 performs a process of counting the number of people or groups moving from top to bottom along the measurement boundary line A1.

(C) of FIG. 6 shows an example of processing in the second layer. In the second layer, the information processing device 2 performs a process of counting the number of people or groups moving from bottom to top along the measurement boundary line A1.

In this way, the information processing device 2 separates the process of counting the number of people based on the direction of movement of the person or group. This allows the information processing device 2 to appropriately count the number of people passing through the measurement boundary line A1.

Figure 7 is a diagram showing an example block configuration of the information processing device 2 and the server 3. The radar device 1 is also shown in Figure 7. As shown in Figure 7, the information processing device 2 has a setting unit 11, an information correction unit 12, a grouping unit 13, a discrimination unit 14, a detection memory unit 15, a learning information memory unit 16, a discrimination information learning unit 17, a number of people estimation unit 18, and a passage determination unit 19. The server 3 has an information analysis unit 21 and a display unit 22.

The setting unit 11 receives input of information on the installation state of the radar device 1, such as the inclination, and road conditions measured by the radar device 1. Information on people flow and vehicular traffic volume is also fed back to the setting unit 11 from the server 3. The setting unit 11 generates correction information based on the information on the installation state of the radar device 1, the information on road conditions measured by the radar device 1, and the information fed back from the server 3. The setting unit 11 outputs the generated correction information to the information correction unit 12.

The information correction unit 12 receives the point cloud data output from the radar device 1. The information correction unit 12 corrects the received point cloud data based on the correction information output from the setting unit 11. For example, the information correction unit 12 corrects the point cloud data of a person moving in a place where it is difficult for the radar device 1 to capture point cloud data based on the correction information so that the point cloud data becomes clear.

The grouping unit 13 groups the corrected point cloud data, for example, based on the density of the point cloud data.

The discrimination unit 14 discriminates whether the grouped point cloud data represents a person or a vehicle (primary discrimination). The discrimination unit 14 discriminates whether the grouped point cloud data represents a person or a vehicle based on the learning information stored in the learning information storage unit 16. The discrimination unit 14 estimates (obtains) the number of people in the group from the area of the group and the number of point clouds in the group. The discrimination unit 14 stores the primary discrimination result in the detection storage unit 15.

The grouping unit 13 regroups the grouped point cloud data based on the primary discrimination result stored in the detection storage unit 15 and the secondary discrimination result described later. For example, the grouping unit 13 executes the regrouping process described in FIG. 4B.

The discrimination unit 14 discriminates whether the regrouped group is a person or a vehicle (secondary discrimination). The discrimination unit 14 performs the secondary discrimination based on the learning information stored in the learning information storage unit 16. The discrimination unit 14 stores the secondary discrimination result in the detection storage unit 15.

The detection memory unit 15 stores the primary and secondary discrimination results of the discrimination unit 14. The primary and secondary discrimination results include detection records (ID, number of people, previous position, current position, and relative position change).

The learning information storage unit 16 stores the learning information learned by the discrimination information learning unit 17 described below.

The discrimination information learning unit 17 learns in advance information such as area, speed, and gap amount required to discriminate people, vehicles, and regroup people, and generates learning information. The generated learning information is stored in the learning information storage unit 16.

The number of people estimation unit 18 estimates (determines) the number of people in the regrouped groups and the number of people in the separated groups based on the secondary discrimination result of the discrimination unit 14.

The passage determination unit 19 determines the passage of people and vehicles across the measurement boundary line A1 based on the relative position change of the detection record. When the passage determination unit 19 determines the passage of people and vehicles across the measurement boundary line A1, it counts the number of people and vehicles passing through the measurement boundary line A1. For the regrouped groups that pass through the measurement boundary line A1, the passage determination unit 19 counts using the number of people estimated by the number of people estimation unit 18. The passage determination unit 19 transmits the counted number of people and vehicles to the server 3.

The information analysis unit 21 of the server 3 analyzes the number of people and vehicles transmitted from the information processing device 2. For example, the information analysis unit 21 calculates the cumulative number of people and vehicles that pass through the measurement boundary line A1 per unit time.

The display unit 22 displays the analysis results of the information analysis unit 21 on a display device.

The detection memory unit 15 and learning information memory unit 16 shown in FIG. 7 may be configured with a memory. The functions of the other blocks shown in FIG. 7 may be realized by a processor such as a CPU (Central Processing Unit). The processor executes a program stored in the memory to realize the functions of the other blocks shown in FIG. 7.

The grouping unit 13 may separately execute a grouping process of people moving from a first direction toward the measurement boundary line A1 and a grouping process of people moving from a second direction opposite to the first direction toward the measurement boundary line A1. The passage determination unit 19 may separately execute a counting process of people moving from the first direction toward the measurement boundary line A1 and a counting process of people moving from the second direction toward the measurement boundary line A1.

The block configuration of the information processing device 2 is not limited to the example of FIG. 7. For example, the grouping unit 13 may have the functions of the discrimination unit 14. The grouping unit 13 may have the functions of the discrimination unit 14 and the number of people estimation unit 18.

FIGS. 8A and 8B are flowcharts showing an example of processing by the information processing device 2. The information processing device 2 periodically and repeatedly executes the flowcharts shown in FIG. 8A and FIG. 8B. The information processing device 2 may execute the flowcharts shown in FIG. 8A and FIG. 8B independently in each of the first layer processing and the second layer processing.

The information processing device 2 generates correction information based on information on the installation state of the radar device 1, information on road conditions measured by the radar device 1, and information fed back from the server 3 (S1).

The information processing device 2 receives the point cloud data from the radar device 1 (S2).

The information processing device 2 corrects the point cloud data received in S2 based on the correction information generated in S1 (S3).

The information processing device 2 groups the point cloud data corrected in S3, for example, based on the density of the point cloud data (S4).

Based on the learning information, the information processing device 2 initially determines whether the grouped point cloud data represents a person or a vehicle (S5).

The information processing device 2 determines whether there are other groups within a short distance (within a specified distance) from a certain group (S6).

When the information processing device 2 determines that no other groups exist within a predetermined distance from a certain group ("No other groups exist" in S6), the process proceeds to S8.

On the other hand, if the information processing device 2 determines that another group exists within a predetermined distance from a certain group ("Another group exists" in S6), it regroups the certain group with the other group (S7).

The information processing device 2 acquires the detection record from the detection memory unit 15 (S8). Note that the acquired detection record includes the location information and number of people.

The information processing device 2 performs a secondary discrimination of the grouped groups and the regrouped groups based on the detection record and learning information acquired in S8 (S9). When the information processing device 2 determines that the group subjected to the secondary discrimination is a vehicle ("Vehicle" in S9), the process proceeds to S13.

On the other hand, if the information processing device 2 determines that the secondary determined group is a person ("Person" in S9), it determines whether the group is a regrouped group (S10).

When the information processing device 2 determines in S10 that the group is a regrouped group (regrouping target in S10), it performs a process of handing over the number of people (S11). For example, as described in FIG. 4B, the number of people in the detection record of the group immediately before separation is assigned to the separated group and carried over to the current detection record.

If the information processing device 2 determines in S10 that the group is not a regrouped group (not subject to regrouping in S10), it estimates the number of people not subject to regrouping (S12).

The information processing device 2 stores the detection record in the detection memory unit 15 (S13). The detection record includes the number of people assigned in S11 and the number of people estimated in S12.

The information processing device 2 determines whether people and vehicles have passed through the measurement boundary line A1 (S14). If the information processing device 2 determines that people and vehicles have not passed through the measurement boundary line A1 (no passing in S14), it ends the processing of the flowchart.

On the other hand, if the information processing device 2 determines that people and vehicles have passed through the measurement boundary line A1, it counts the number of people and vehicles that have passed through the measurement boundary line A1 (S15).

The information processing device 2 transmits the number of people and the number of vehicles counted in S15 to the server 3 (S16). Then, the information processing device 2 ends the processing of this flowchart.

As described above, the information processing device 2 has a grouping unit 13 that groups point cloud data measured by the radar device 1 that emits radio waves toward the passageway and links the number of people included in the group to the group, and a passing determination unit 19 that counts the number of people linked to the group who have passed the measurement boundary line A1 set on the passageway. This allows the information processing device 2 to simply and appropriately count the number of people who have passed the measurement boundary line set on the passageway using radio waves.

For example, the information processing device 2 identifies people moving through a passageway based on point cloud data measured by the radar device 1 using radio waves that are not dependent on brightness. The information processing device 2 tracks crowded people as a group, rather than tracking them individually. Then, when the group passes through the measurement boundary line A1, the information processing device 2 counts the number of people associated with the group as the number of people who have passed through the measurement boundary line A1. This allows the information processing device 2 to simply and appropriately count the number of people who have passed through the measurement boundary line A1 using radio waves.

In the above, the information processing device 2 groups and counts people, but this is not limited to the above. The information processing device 2 may also group and count any objects, such as vehicles.

In addition, in the above, the information processing device 2 counts people and vehicles, but this is not limited to this. The information processing device 2 may count either people or vehicles.

The counting system is not limited to counting the amount of people passing through roads. For example, the counting system may count the amount of people passing through a specific passageway in a warehouse. The counting system may also analyze the flow of people in the warehouse.

In the above-described embodiments, the notation "part" used for each component may be replaced with other notations such as "circuitry", "device", "unit", or "module".

Although the embodiments have been described above with reference to the drawings, the present disclosure is not limited to such examples. It is clear that a person skilled in the art can conceive of various modified or amended examples within the scope of the claims. It is understood that such modified or amended examples also fall within the technical scope of the present disclosure. Furthermore, the components in the embodiments may be combined in any manner as long as it does not deviate from the spirit of the present disclosure.

The present disclosure can be realized by software, hardware, or software in conjunction with hardware. Each functional block used in the description of the above embodiments may be realized, in part or in whole, as an LSI, which is an integrated circuit, and each process described in the above embodiments may be controlled, in part or in whole, by one LSI or a combination of LSIs. The LSI may be composed of individual chips, or may be composed of one chip that includes some or all of the functional blocks. The LSI may have data input and output. Depending on the degree of integration, the LSI may be called an IC, a system LSI, a super LSI, or an ultra LSI.

The integrated circuit method is not limited to LSI, and may be realized by a dedicated circuit, a general-purpose processor, or a dedicated processor. In addition, a field programmable gate array (FPGA) that can be programmed after LSI manufacturing, or a reconfigurable processor that can reconfigure the connections and settings of circuit cells inside the LSI, may be used. The present disclosure may be realized as digital processing or analog processing.

Furthermore, if an integrated circuit technology that can replace LSIs emerges due to advances in semiconductor technology or other derived technologies, it is natural that such technology can be used to integrate functional blocks. The application of biotechnology, etc. is also a possibility.

This disclosure is useful for information processing devices that count people and vehicles passing through passageways such as roads.

REFERENCE SIGNS LIST 1 Radar device 2 Information processing device 3 Server 4 Network 11 Setting unit 12 Information correction unit 13 Grouping unit 14 Discrimination unit 15 Detection storage unit 16 Learning information storage unit 17 Discrimination information learning unit 18 Number of people estimation unit 19 Passage determination unit 21 Information analysis unit 22 Display unit A1 Measurement boundary line

Claims (6)

a grouping unit that groups point cloud data measured by a radar device that emits radio waves toward the passageway and associates the number of objects included in each group with the group;
a passage determination unit that counts the number of objects associated with the group that have passed through a measurement boundary line set on the passage,
The grouping unit associates a current position of the group and a previous position of the group with the group;
the passage determination unit determines whether the group has passed the measurement boundary line based on a positional relationship between the current position and the previous position with respect to the measurement boundary line,
When the group is a separated group separated from a pre-separation group, the grouping unit links the position of the pre-separation group to the separated group as the previous position.
Information processing device. The grouping unit assigns the number of objects associated with the pre-separation group as the number of objects in the separation group.
The information processing device according to claim 1 . a grouping unit that groups point cloud data measured by a radar device that emits radio waves toward the passageway and associates the number of objects included in each group with the group;
a passage determination unit that counts the number of objects associated with the group that have passed through a measurement boundary line set on the passage,
When the group is a combined group formed by combining a plurality of pre-combined groups, the grouping unit links each position of the plurality of pre-combined groups to the combined group as a previous position .
Information processing device. the grouping unit divides the measurement boundary line into a first grouping processing layer that groups objects moving toward the measurement boundary line from a first direction, and a second grouping processing layer that groups objects moving toward the measurement boundary line from a second direction opposite to the first direction;
the passage determination unit divides the counting process into a first counting process layer that counts objects moving from the first direction toward the measurement boundary line, and a second counting process layer that counts objects moving from the second direction toward the measurement boundary line.
The information processing device according to claim 1 . The radar equipment emits radio waves toward the corridor and collects measured point cloud data, which is then grouped together.
The number of objects included in a group is associated with the group;
Counting the number of objects associated with the group that pass through a measurement boundary line set on the passage;
Linking the current location of the group and the previous location of the group to the group;
determining whether the group has passed through the measurement boundary line based on a positional relationship between the current position and the previous position with respect to the measurement boundary line;
If the group is a separated group separated from a pre-separation group, the position of the pre-separation group is linked to the separated group as the previous position.
Counting method. The radar equipment emits radio waves toward the corridor and collects measured point cloud data, which is then grouped together.
The number of objects included in a group is associated with the group;
Counting the number of objects associated with the group that pass through a measurement boundary line set on the passage;
When the group is a combined group formed by combining a plurality of pre-combined groups, the positions of the plurality of pre-combined groups are linked to the combined group as previous positions.
Counting method.

Images