JP2005215909A - Town traffic state investigation system using moving picture processing technique - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a system determining a traffic volume at a specific point in a town after considering the moving directions by identifying persons, vehicles, and other objects. <P>SOLUTION: A security camera is used as a camera for a traffic volume investigation. Moving picture data photographed with the cameras in the town are transmitted to a center. The center applies moving picture identification processing to moving picture data collected from various places in a batch. Persons, bicycles, motorcycles, cars are identified and each moving direction is extracted to count the traffic volume on an object moving direction-by-object moving direction basis. The image processing result is then displayed in real time on a display device. Furthermore, extracting the movement grasps the flow of the object. Thus, the traffic volume after several minutes or several hours can be forecasted by conducting a simulation from statistical processing for the past data. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention relates to a system for investigating and analyzing traffic volume at a certain point in an urban area using moving image processing technology, and in particular, identifies traffic objects, analyzes movements, and calculates traffic volume for each movement for each object. The present invention relates to a traffic situation survey system for counting.

  2. Description of the Related Art Conventionally, as a system for investigating road traffic volume, a system for investigating the number of vehicles using a surveillance camera installed on a road for the purpose of reducing traffic congestion is known. However, there are still few systems for automatically investigating the traffic volume of people whose shape is complicated and high in comparison with vehicles, and as shown in Patent Document 1, a system that measures the number of people passing by using a ticket gate of a station Is the mainstream. However, in recent years, with the improvement of moving image processing technology, a system as shown in Patent Document 2 has been proposed that can identify and count people and other objects by shape.

JP 2002-297757 A JP 2002-074371 A

  For the purpose of investigating the traffic volume and traffic flow at a specific point in an urban area, it is practiced to identify and measure people, vehicles, and other objects. Such a survey is performed by a person visually identifying an object and counting for each object. However, when automating this, a person passing through a station when the system shown in Patent Document 1 is applied. There was a problem that only measurement was possible. In addition, when a system such as that disclosed in Patent Document 2 is applied, it is possible to accurately detect the object identification and its count, but the focus is on reducing the error in the number of passing people per unit time. However, it does not consider the direction of travel, and it is not suitable for traffic surveys in urban areas where it is desirable to investigate how much people are flowing on a more macro level.

  An object of the present invention is to solve the above-mentioned problems and to provide a system for measuring the traffic volume at a specific point in an urban area in consideration of the moving direction by identifying a person, a vehicle, or another object.

  In order to solve the above-described problems, the traffic situation survey system of the present invention installs a monitoring camera in a place overlooking a city road, and transmits video data reflected on the camera to the traffic situation survey system through a network. The camera used here does not need a special function, and needs only to have a certain resolution, and may be shared with a security camera in the city. In addition, in order to investigate the traffic volume of the entire city area and investigate the flow of people and vehicles, surveillance cameras are installed at multiple points in the city, and the number from each camera is numbered. To the traffic condition survey system. The video data is transmitted as a still image of several frames or a moving image of a few seconds at regular intervals, and is analyzed by an image identification processing unit provided in the traffic condition investigation system. The image identification processing unit has a function of identifying an object based on the feature amount of the image, and extracts an image suitable for an object to be investigated such as a person, a bicycle, a motorcycle, or an automobile. Furthermore, the moving direction of each object is also extracted by a plurality of still image changes or moving image movements, and each object is counted by the image counting unit for each moving direction. The counting result is accumulated in the counting information accumulating unit together with the time information and used as basic data for traffic analysis. Further, the traffic condition survey system may include a simulation unit, and may predict the traffic volume at a future time by using past data stored in the count information storage unit. In addition, the display of the traffic situation survey system links the camera position and the address on the map, and provides a function to select and display the traffic volume for each object. A function of displaying the degree of congestion of the object in different colors may be provided. Furthermore, the display unit of the traffic condition investigation system may be provided with a function of selecting a point on the map and displaying the traffic volume at the point as a numerical value or a graph. The counting results are stored in association with conditions such as weather, time, etc., and by inputting the conditions, a simulation is performed from past data stored in the database, and a function of predicting the traffic volume at the future time of the day is provided. Also good.

  As described above, according to the present invention, traffic data is transmitted to a traffic condition survey system using a camera installed in an urban area, and counting is performed for each object and direction using a moving image processing technique for identifying an object by the traffic condition survey system. As a result, it is possible to easily measure the traffic volume on a city road where objects are mixed. Moreover, by collecting traffic data at a plurality of points according to movement, it is possible to grasp the traffic flow of the entire city area. Furthermore, traffic volume data is continuously collected and stored in association with conditions such as day of the week, season, time zone, weather, etc., so that traffic volume and traffic flow at a specific point under certain conditions can be predicted. It becomes possible.

  Hereinafter, the best mode for carrying out the traffic condition investigation system of the present invention will be described with reference to the drawings. FIG. 1 is a diagram showing a configuration example of a traffic condition investigation system of the present invention. 1 is a camera installed at a plurality of points, 2 is a network for transmitting data, and 3 is a traffic condition survey system. The traffic condition investigation system 3 includes an image data receiving unit 4, an image identification processing unit 5, an image counting unit 6, a counting information storage unit 7, and a display unit 9, and preferably includes a simulation unit 8. Here, it is assumed that a moving image is transmitted as traffic data for several seconds or several minutes.

  FIG. 2 shows the flow of traffic volume survey processing by the traffic situation survey system of the present invention. The camera 1 is installed on a plurality of roads in the city, and the video data reflected on the camera is transmitted to the traffic condition investigation system 3 through the network 2 (step 201). In this embodiment, since it is assumed that the cameras are installed at a plurality of points, each camera assigns an identification number to each image and transmits it to the image data receiving unit 4 of the traffic condition investigation system 3. To do. The moving image data received by the image data receiving unit 4 is identified by the image identification processing unit 5 as a person, a bicycle, a motorcycle, and a car, and the movement of each object is also identified (step 202). In the captured video, the image counting unit 6 counts according to the moving direction of each object such as descending when the counting object moves to the right and ascending when it moves to the left (step 203). The count result is stored in the count information storage unit 7 together with the time information when the video is received (step 204). Traffic volume can be predicted by the simulation unit 8 using past data stored in the count information storage unit 7. When the simulation condition is input and the simulation button is pressed (step 205), the simulation unit 8 predicts the traffic volume at the future time (step 206) and displays it on the display unit 9 (step 207).

  An example of the data configuration of the count information storage unit 7 is shown in FIG. FIG. 3 is a database in which the camera position and the acquisition date and time are associated with each other and counted separately for each object. 4 shows a specific name of the object code, and FIG. 5 shows a specific place of the position code. Since the data of the counting information storage unit 7 is linked to the camera position and the address on the map, the traffic at the target position is displayed by displaying the map on the display unit 6 and clicking the position on the map. be able to. FIG. 6 shows an example of the display screen. As shown in FIG. 6, when the button 10 indicating the traffic volume of the person is clicked on the display of the display unit 9 (step 208), the congestion level of the person is displayed in a color-coded manner on the map of the entire city (step 209). ). Similarly, when the button 11 indicating the bicycle traffic is clicked, the degree of congestion of the bicycle is displayed in different colors on the map of the entire city. Bikes and cars are displayed in the same way. More specifically, FIG. 7 shows an example of a display screen when the count number is displayed. When the point C 12 is clicked on the map of FIG. 7, the number of counts for each point and the number of uplinks and downlinks is displayed as a graph as shown in FIGS. In this way, the traffic volume can be grasped at a glance on the map.

  The traffic volume investigated as described above can be used not only for simple simulation and traffic volume for each point, but also for more advanced analysis. For example, consider the necessity of providing a sidewalk or a parking lot in a store based on a comparison of the traffic volume of people and cars, or if there is a lot of traffic in one direction, increasing the lane on one side only, etc. It can be used as data to guide city planning guidelines from the perspective of more detailed traffic development.

  It can be said that the traffic volume of the local government can be grasped at a glance for people, bicycles, motorcycles, cars, etc., and there is a possibility that it will be useful information for companies such as setting up stores on roads where many people pass.

It is the schematic which shows one Embodiment of the traffic condition investigation system by this invention. It is a flowchart which shows the process of the traffic condition investigation system by this invention. It is the database record block diagram which matched the camera position and acquisition date with each object of the traffic condition investigation system by this invention, and counted according to going up and down. It is a database record block diagram which shows the specific name of the object code of the traffic condition investigation system by this invention. It is a database record block diagram which shows the specific place of the position code of the traffic condition investigation system by this invention. It is a figure which shows the traffic condition display part of the whole traffic condition investigation system by this invention. It is a figure which shows the up-and-down count number display part according to the point of the traffic condition investigation system by this invention. It is a figure which shows the up-and-down traffic volume display part for every area of the traffic condition investigation system by this invention. It is a figure which shows the traffic amount prediction display part classified by area of the traffic condition investigation system by this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 ... Camera installation location in FIG. 1, 2 ... Network part in FIG. 1, 3 ... Traffic situation investigation system in FIG. 4, 4 ... Image data receiving part in FIG. 1, 5 ... Image identification processing part in FIG. 1, 7... Counting information storage unit in FIG. 1, 8... Simulation unit in FIG. 1, 9. Display unit in FIG. 10, 10. Button indicating the number of bicycles to be counted, 12 ... C point in FIG.

Claims (3)

  In a traffic situation survey system in a city using images acquired by a surveillance camera, the traffic situation survey system is equipped with a camera that captures the traffic situation at a specific point and transmits it to the traffic situation survey system. An image counting unit and a counting information storage unit, and the image identification processing unit has a function of identifying an object from a plurality of still images or moving images taken at fixed time intervals and a function of identifying the moving direction of the object, A traffic situation survey system for a city using an image processing technique characterized in that a function for counting each object and each moving direction of the object is provided in the section.   2. The urban traffic condition investigation system according to claim 1, wherein the counting information storage unit is associated with at least one of a weather condition, a date, and an event in addition to at least a video acquisition time or a video acquisition time, and each object and the direction in which the object moves. Each count result is stored, the traffic condition investigation system has a simulation unit, and has a function to simulate the traffic condition according to the object after several hours based on the information of the count information accumulation unit A traffic situation survey system in a city using the image processing technology according to claim 1. 3. The traffic situation survey system for a city according to claim 1 or 2, wherein the traffic situation survey system comprises map information, information associated with the map position and the camera position, and a map is displayed on the display section of the traffic situation survey system. The city traffic situation survey system using the image processing technology according to claim 1 or 2, further comprising a function of displaying a camera position and traffic information for each camera in association with each other.

Images