Sensor Placement Determination Device and Sensor Placement Determination Method

Patent metadata

Field	Value
Publication number	US-2016142679-A1
Application number	US-201314899834-A
Country	US
Kind code	A1
Filing date	Jun 21, 2013
Priority date	Jun 21, 2013
Publication date	May 19, 2016
Grant date	—

Abstract

This sensor placement determination device is provided with an external storage device ( 15 ) in which traffic data, i.e. the actual paths taken by people, is recorded in a planned target area in which sensors are to be arranged, a traffic simulation observation unit ( 100 ) which, for candidate arrangement positions of the sensors in the target area, generates, on the basis of the traffic data, information about the passage of people through the monitoring area of the sensors, a traffic estimation unit ( 102 ) which, with the passage information as a constraining condition, generates candidate paths by which people might pass through the target area, an estimated error calculation unit ( 104 ) which, for the traffic through the monitoring area of the sensors, calculates an estimated error between the number of people passing along the candidate paths and the number of people passing along the actual paths, and an additional camera determination unit ( 106 ) which determines a sensor arrangement position that minimizes the calculated estimated traffic error.

First claim

1 . A sensor placement determination device comprising:
storage means configured to store, about a target area where a plurality of sensors that observes traffic is to be placed, traffic data that is an actual path of when a person actually passes through the target area; traffic simulation measurement means configured to simulatively generate observation information of the person observed with each of the sensors, when the plurality of sensors is placed, based on the traffic data, in candidates of placement positions of the plurality of sensors in the target area; traffic estimation means configured to estimate and generate a candidate path indicating how the person passes through the target area, using the observation information generated by the traffic simulation measurement means as a constraining condition; estimated error calculation means configured to compare the number of observed people of each candidate of the placement positions of the sensors calculated based on the candidate path generated by the traffic estimation means and the number of observed people of each candidate of the placement positions of the sensors calculated based on the traffic data, and calculate an estimated error of when the traffic estimation means estimates the candidate path; and sensor position determination means configured to select the candidate as the placement position of the sensor where the estimated error calculated by the estimated error calculation means becomes small, for the candidates of the placement positions of the plurality of sensors, and determines the placement position.

2 . The sensor placement determination device according to claim 1 , wherein
the traffic estimation means generates the candidate path by multi-agent simulation.

3 . The sensor placement determination device according to claim 2 , wherein
the traffic estimation means generates the candidate path matched with an observation value that is the observation information of the person observed with each of the sensors by a data assimilation technique, by weighting the candidate path with the observation value.

4 . The sensor placement determination device according to claim 3 , wherein
the estimated error calculation means calculates the number of observed people for each area of the candidates of the placement positions of the sensors, of the target area, and uses a root mean square error of the number of people of each area, as the estimated error.

5 . The sensor placement determination device according to claim 1 , wherein
the sensor is at least one of a camera, a line sensor, and a laser scanner.

6 . The sensor placement determination device according to claim 1 , wherein
the sensor placement determination means selects the candidate of the placement position of the sensor for each passage portion where divergence does not exist.

7 . The sensor placement determination device according to claim 1 , wherein
the sensor placement determination means displays the target area and the determined placement position of the sensor in a display device.

8 . A sensor placement determination method of adding a sensor in a target area, using a sensor placement determination device including a storage means that stores, about the target area where a plurality of sensors that measures traffic is to be placed, traffic data that is an actual path of when a person actually passes through the target area, traffic simulation measurement means, traffic estimation means, estimated error calculation means, and sensor position determination means,
the traffic simulation measurement means simulatively generating observation information of the person observed with each of the sensors, when the plurality of sensors is placed, based on the traffic data, in candidates of placement positions of the plurality of sensors in the target area; the traffic estimation means estimating and generating a candidate path indicating how the person passes through the target area, using the observation information generated by the traffic simulation measurement means as a constraining condition; the estimated error calculation means comparing the number of observed people of each candidate of the placement positions of the sensors calculated based on the candidate path generated by the traffic estimation means and the number of observed people of each candidate of the placement positions of the sensors calculated based on the traffic data, and calculating an estimated error of when the traffic estimation means estimates the candidate path; and the sensor position determination means selecting the candidate as the placement position of the sensor where the estimated error calculated by the estimated error calculation means becomes small, for the candidates of the placement positions of the plurality of sensors, and determining the placement position.

Assignees

• Hitachi Ltd

Inventors

• Miyoshi Masanori
• Katou Manabu

Classifications

• G06K9/00778

  Physics · mapped topic

• H04N7/181 Primary

  for receiving images from a plurality of remote sources · CPC title

• H04N7/18 Primary

  Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast · CPC title

• G06V20/53

  Recognition of crowd images, e.g. recognition of crowd congestion · CPC title