KR101735557B1 - System and Method for Collecting Traffic Information Using Real time Object Detection - Google Patents
System and Method for Collecting Traffic Information Using Real time Object Detection Download PDFInfo
- Publication number
- KR101735557B1 KR101735557B1 KR1020150138639A KR20150138639A KR101735557B1 KR 101735557 B1 KR101735557 B1 KR 101735557B1 KR 1020150138639 A KR1020150138639 A KR 1020150138639A KR 20150138639 A KR20150138639 A KR 20150138639A KR 101735557 B1 KR101735557 B1 KR 101735557B1
- Authority
- KR
- South Korea
- Prior art keywords
- vehicle
- traffic information
- image processing
- region
- interest
- Prior art date
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 51
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012545 processing Methods 0.000 claims description 59
- 230000001133 acceleration Effects 0.000 claims description 14
- 230000005540 biological transmission Effects 0.000 claims description 10
- 238000012937 correction Methods 0.000 claims description 5
- 238000007689 inspection Methods 0.000 claims description 5
- 238000010295 mobile communication Methods 0.000 claims description 5
- 238000010191 image analysis Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 2
- 238000003672 processing method Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 description 17
- 238000010586 diagram Methods 0.000 description 12
- 238000004891 communication Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000010365 information processing Effects 0.000 description 2
- HPTJABJPZMULFH-UHFFFAOYSA-N 12-[(Cyclohexylcarbamoyl)amino]dodecanoic acid Chemical compound OC(=O)CCCCCCCCCCCNC(=O)NC1CCCCC1 HPTJABJPZMULFH-UHFFFAOYSA-N 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000010420 art technique Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000011514 reflex Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Images
Classifications
-
- G06Q50/30—
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
-
- H04N5/2257—
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention relates to a traffic information collection system and method by real time target detection using a real time target detection method for changing a region of interest dynamically and adjusting a frame to be processed in an input image, A smart camera for a vehicle that collects traffic information in real time through video information and sensor information of the camera and transmits the collected traffic information to a traffic information providing server through a network, analyzes the traffic information transmitted from the smart camera of the vehicle, And a traffic information providing server for providing a real time traffic information service about the degree of congestion of the vehicle. The smart camera of the vehicle dynamically changes a region of interest and adjusts frames to be processed in the input image to detect real time targets.
Description
The present invention relates to an advanced driver assistance system and a traffic information collection system, and more particularly, to a traffic information collection system by real time target detection using a real time target detection method of dynamically changing a region of interest and adjusting a frame to be processed in an input image, ≪ / RTI >
Recently, smartcar technology has been actively studied due to the increase of social necessity for safety and convenience in driving and development of semiconductor technology and mobile communication technology.
Smart car technology is a technology that provides driver safety management and road information by connecting sensor and wireless communication device to vehicle environment. It can be classified into car camera technology, traffic volume management technology, and road view service.
1 is a block diagram showing an example of a general advanced driver assistance system.
Automotive camera technology is represented by automotive black boxes and Advanced Driver Assistant System (ADAS).
The vehicle black box is a device that stores driving images and identifies the cause of the accident through images around the accident point when an accident occurs. Recent developments in related technologies have enabled multi-channel, high-resolution recording. In addition, various communication modules can be used to receive incident images via wireless communication, or in real time to store vehicle information in conjunction with on-board diagnostics (OBD).
Advanced driver assistance system is a system that assists the driver to drive safely through lane, surrounding vehicle and pedestrian recognition by utilizing radar, camera and various sensors mounted on vehicle.
Systems included in these advanced driver assistance systems include safety related systems such as the Lane Departure Warning System (LDWS) and the Adoptive Cruise Control (ACC).
Traffic volume management technologies include the Intelligent Transport System and the Transport Protocol Experts Group (TPEG).
Intelligent transportation system means transportation system that improves traffic efficiency and safety by developing and utilizing advanced transportation technology and traffic information such as electronic control and communication in transportation and traffic facilities.
Intelligent traffic systems that can be encountered in real life include bus arrival guidance system, automatic intersection signal change system, and real time traffic information service provided through navigation.
At this time, a teacack is mainly used as a method for providing real-time traffic volume information to navigation. TiePack is a technology that provides real-time traffic information and travel information through DMB (Digital Multimedia Broadcasting) broadcasting network.
Since it is a technology that needs to utilize the DMB broadcasting network, existing domestic airwaves and terrestrial broadcasting companies are serving. The TIPPACK provides five services: traffic congestion information, safe driving information, evacuation information, news information, and interest information.
The Road View service is a service that provides real-distance images of each area through a digital single lens reflex (DSLR) camera and high-resolution panorama photographs. Currently, Rod View service is mainly offered in large portal sites such as Daum, Naver, and Google, and it has a function to zoom in and out and look forward.
In the prior art, a target detection method for an advanced driver assistance system is as follows.
In the prior art, a method of dividing a fixed region of interest into two or more regions is used in order to set a region of interest with a fixed size for detecting a target in an advanced driver assistance system environment, i.e., an embedded system environment. In this case, the region of interest refers to a region in which an operation of a computer vision algorithm is performed on an input image.
A prior art technique for setting the area of interest at fixed size in advanced driver assistance systems is to exclude unneeded portions such as sky or guidance when recognizing pedestrians in front of the vehicle.
In recognition of the license plate, two different fixed areas of interest are used, and different algorithms operating in each area of interest are used to recognize the license plate of the vehicle in close proximity to the vehicle in the far distance. There is a way.
A method and an apparatus for collecting real-time traffic information according to the related art will now be described.
The prior art focuses on collecting traffic information in real time using various vehicle speeds such as the passing speed of each vehicle and the speed of the vehicle calculated using GPS.
In order to measure the passage speed of each section, a position transmitting device is installed at the start and end points of each section, and a receiving device capable of collecting signals from the position transmitter is installed in a position receiver And calculates the speed using the received location information and time and obtains traffic flow information through the calculated speed.
Prior art utilizing GPS utilizes GPS to obtain the position of the vehicle in real time, calculates the change in position per unit time, and calculates the speed of the vehicle. Based on the calculated speed of the vehicle, the degree of congestion for each road segment is determined.
However, the target detection method and the real-time traffic information collection method and apparatus of the related arts are limited in terms of reducing the traffic congestion cost of the whole society because they are operated individually or are difficult to construct an integrated system.
The present invention solves the problems of the prior art driver assistance system and traffic information collection system of the related art. The present invention provides a real-time target detection method using a real-time target detection method of dynamically changing a region of interest and adjusting a frame to be processed in an input image The present invention relates to a traffic information collection system and method.
The present invention provides a method and system for collecting traffic information by real-time target detection having a real-time target detection configuration that dynamically changes a frame to be processed in an input image according to a road situation and a method of dynamically changing a region of interest It has its purpose.
The present invention provides a system and method for collecting traffic information by real-time target detection having a configuration for collecting traffic information in real time using the speed of the vehicle, the distance from the head vehicle, .
The present invention relates to a real-time target detection technique for an advanced driver assistance system and a real-time traffic information collecting device and method using the same, thereby providing safety and convenience for the user and reducing the traffic congestion cost of the entire society And a traffic information collection system and method using real-time target detection.
The objects of the present invention are not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood by those skilled in the art from the following description.
In order to achieve the above object, according to the present invention, there is provided a traffic information collection system using real-time target detection, which collects traffic information in real time through vehicle information, image information of a camera, and sensor information, A traffic information providing server for analyzing traffic information transmitted from a smart camera for a vehicle and providing a user with a real time traffic information service about a photograph of a current road and a degree of congestion; And a smart camera for a vehicle is characterized by dynamically changing a region of interest and adjusting a frame to be processed in an input image to detect a real time target.
Here, in order to dynamically change the area of interest, the smart camera for vehicles selects the number of frames to be processed among consecutive frames input based on the speed and the amount of change in acceleration of the vehicle, and the number of frames (FPF)
In order to change the area of interest dynamically, the smart camera of the vehicle extracts a fixed area corresponding to the character of each target in the entire input image, designates the area around the target detected in the fixed area of interest as the area of interest, The target region is detected while moving the region of interest.
The smart camera of the present invention is characterized by measuring the speed of the vehicle, the distance from the leading vehicle, and the number of the preceding vehicle when measuring the traffic volume for real time traffic information collection.
The smart camera for a vehicle includes an AP (Application Processor) for real-time image processing, a coprocessor for controlling the sensors, a sensor used for acquiring peripheral information of the vehicle, an acceleration sensor, and a GPS (Global Positioning System) A mobile communication network module for transmitting collected traffic information to a traffic information providing server, a Wi-Fi module for informing a driver of a dangerous situation through a driver's smart phone, and an in-vehicle ECU to acquire vehicle status information And an OBD scanner for the OBD scanner.
The smart camera for vehicles is an image processing program that detects a target object from an image and stores it in a database, a media transmission program that inquires a list of still images in a database, transmits a media file that has not been transmitted to a traffic information providing server, And a data transmission program for transmission of the general data except for the data transmission.
In addition, the smart camera of the vehicle analyzes the surrounding situation in real time through image recognition, informs the driver of the image analysis result made of the image metadata, stores it in the database, periodically samples the sensor values including the OBD, And stores it in a database, and combines the image metadata and the sensor data into one, and transmits the combined image metadata to the traffic information server.
The smart camera of the vehicle recognizes the number plate of the front vehicle by analyzing the collected image to measure the distance from the leading vehicle, estimates the distance from the front vehicle based on the position of the number plate recognized in the input image, Converting the location to the distance to the front vehicle
The traffic information providing server transmits the traffic volume information based on the speed of the vehicle, the distance to the head vehicle, and the number of the preceding vehicles,
According to another aspect of the present invention, there is provided a traffic information collection system using real-time target detection, wherein the traffic information collection system dynamically changes a region of interest and adjusts a frame to be processed in an input image, A frame checking unit for checking whether a received frame is an object to be examined, an image processing region determining unit for determining an image processing region to separately perform the fixed region image processing or the dynamic region image processing, the fixed region image processing or the dynamic region image processing, A target detection unit for detecting a target through a target region setting and recognition, and a target position storage unit for storing a position of a target obtained as a result of the image processing in a rectangular form, .
In order to accomplish another object, there is provided a method for collecting traffic information by real-time target detection according to the present invention, in which, in order to dynamically change a region of interest and adjust a frame to be processed in an input image, Determining whether the object is an inspection object, determining whether the object is a dynamic region in order to distinguish the fixed region image processing or the dynamic region image processing, performing a fixed region image processing when it is determined that the object is not a dynamic region, A step of performing a dynamic region image processing, a step of performing a fixed region image processing and a dynamic region image processing, and storing the position of the object obtained as a result of the image processing in a rectangular form if an object is found .
Here, the fixed area image processing includes recognizing a pedestrian, a front license plate, a front sign, and a lane, and setting an area of interest fixed in a position and a size according to characteristics of each object.
In the dynamic region image processing, a region of interest is set in consideration of the movement of an object on the basis of a stored target position, and a region of interest is set by extending the smallest rectangle including the recognized object to a predetermined size .
The system and method for collecting traffic information by real-time target detection according to the present invention have the following effects.
First, it can improve traffic information collection performance by using real - time target detection method that dynamically changes the region of interest and adjusts frames to be processed in the input image.
Second, the amount of computation can be reduced by changing the region of interest dynamically and real - time target detection configuration that dynamically adjusts frames to be processed in the input image according to the road conditions.
Third, accuracy of traffic information can be improved by collecting traffic information in real time using the speed of the vehicle, the distance from the leading vehicle, and the number of vehicles in front of the vehicle.
Fourth, real - time target detection technology for advanced driver assistance system and vehicle technology related to real - time traffic information collection device and method utilizing it can be fused to provide safety and convenience of users and reduce traffic congestion cost of society as a whole.
1 is a block diagram showing an example of a general advanced driver assistance system
2 is a block diagram illustrating a real-time target detection method for an advanced driver assistance system according to the present invention.
FIG. 3 is a diagram showing a screen configuration in a traffic information collection system by real-time target detection according to the present invention
FIG. 4 is a hardware configuration diagram of a smart camera of a traffic information collection system by real-time target detection according to the present invention
5 is a software configuration diagram of a smart camera of a traffic information collection system by real-time target detection according to the present invention
6 is a block diagram of a traffic information collection system based on real-time target detection according to the present invention
7 is a flowchart showing a traffic information collection method by real-time target detection according to the present invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a preferred embodiment of a system and method for collecting traffic information by real-time target detection according to the present invention will be described in detail.
The features and advantages of the system and method for collecting traffic information by real-time target detection according to the present invention will be apparent from the following detailed description of each embodiment.
FIG. 2 is a block diagram illustrating a real-time target detection method for an advanced driver assistance system according to the present invention, and FIG. 3 is a screen configuration diagram in a traffic information collection system by real-time target detection according to the present invention.
The present invention has a method of dynamically changing a region of interest to reduce the amount of computation and a real-time target detection scheme of dynamically adjusting a frame to be processed in an input image according to road conditions.
The system and method for collecting traffic information by real-time target detection according to the present invention are largely composed of a smart camera for a vehicle and a traffic information providing server for real-time traffic information service using the same.
Vehicle smart camera collects traffic information in real time through vehicle information, camera image information and other sensor information acquired by using onboard diagnostic system. Then, the collected traffic information is transmitted to the traffic information providing server through the network.
The traffic information providing server analyzes the traffic information received and provides the user with a real time traffic information service, whereby the user can receive the photograph of the current road and the degree of congestion.
The present invention comprises an image processing configuration for detecting a target in an embedded environment and a real-time traffic information collector using the image processing configuration.
The real-time target detection configuration for the advanced driver assistance system applied to the present invention will be described in detail as follows.
In general, a target detection algorithm through popularized image processing requires a large amount of computation. In an embedded environment such as a high-tech driver protection system environment, the amount of computation must be reduced in order to detect targets in real time.
The present invention has a configuration for dynamically changing a region of interest to reduce the amount of computation and a structure for dynamically adjusting frames to be processed in the input image according to road conditions.
First, a configuration for dynamically adjusting a frame to be processed in the input image selects the number of frames to be processed among consecutive frames input based on the vehicle speed and the amount of change in acceleration.
The number of frames (FPF) to be calculated at this time is calculated by the equation in Fig. If FPF is 3, image processing is performed once every 3 frames.
Here, S is the speed of the vehicle,? Is the fine impact correction coefficient,? A is the acceleration change amount, and FR is the frame rate of the target value.
As shown in FIG. 3, the first step is to extract a fixed area corresponding to the characteristics of each target in the entire input image.
The second step is to designate the area around the target detected in the fixed area of interest as the area of interest and detect the target while moving the area of interest according to the change of the position of the target.
The configuration of real - time traffic information collection is as follows.
When measuring road traffic volume, we use 3 kinds of information such as the speed of the vehicle, the distance from the head vehicle, and the number of vehicles in front to more accurately measure the traffic volume and configure the device for this operation.
FIG. 4 is a hardware configuration diagram of a smart camera of a traffic information collection system by real-time target detection according to the present invention, and FIG. 5 is a software configuration diagram of a smart camera of a traffic information collection system by real-time target detection according to the present invention.
The smart camera combines the forward image data and the sensor data acquired through the camera to collect the real-time traffic information to generate the metadata.
The generated metadata is transmitted to the traffic information providing server. The concrete structure of the smart camera required to generate and transmit the metadata is as follows.
Looking at the hardware structure of the smart camera, the smart camera is composed of two processors and various modules as shown in FIG.
The smart camera processor consists of a high performance AP (Application Processor) for real time image processing and a coprocessor for controlling sensors. Other modules included in the device can be divided into a sensor unit and a communication unit.
The sensor unit consists of a camera, an acceleration sensor and a Global Positioning System (GPS) module, which are used to acquire vehicle perimeter information.
Through the camera, it is possible to acquire the forward image and recognize the impact of the vehicle through the acceleration sensor. Finally, the current position and speed of the vehicle can be determined by GPS.
The communication unit is composed of a mobile communication network module, a Wi-Fi module and an OBD scanner.
The traffic information collected using the mobile communication module can be transmitted to the traffic information providing server, and the driver can inform the driver of the danger through the smart phone using the Wi-Fi module. The OBD scanner can also communicate with the in-vehicle ECU to obtain vehicle status information.
As shown in FIG. 5, the library used in the smart camera uses an embedded database, a database connector, and a library (wpa_supplicant) for storing information efficiently in an embedded environment . It also uses a library for real-time image processing (OpenCV, CUDA).
Three applications run in this environment.
The first corresponds to the image processing program of Fig. 5 as the image processing program. This program detects the target object from the image received from the camera and stores it in the database.
The second corresponds to the media transfer program of FIG. 5 as the media transfer program. This program queries the database for a list of still images and transmits media files that have not been transmitted to the server.
The third program is a general data transmission program similar to the media transmission program and corresponds to the data transfer program of FIG.
This program transmits the necessary data from the database to the server in the database.
The operation of such a smart camera is largely composed of three kinds.
First, analyze the surrounding situation in real time through image recognition. The smart camera informs the driver of the image analysis result and stores it in the database. The result of the analysis is called the image metadata.
Second, sensor data including OBD are periodically sampled to generate sensor data and store it in a database.
Third, data stored in the database is transmitted to the traffic information server. When data is transmitted to the server, the image metadata and the sensor data are combined and transmitted.
Data that combines one image metadata and one sensor data is called metadata. There are many kinds of image metadata, but sensor data stores all the sensor data existing in the smart camera as one group, so there are as many types of metadata as the kinds of image metadata.
The image metadata includes pedestrian recognition, license plate recognition, lane departure recognition results, and still images stored periodically. We try to recognize periodically considering the limited amount of computation in the embedded environment.
Basically, the pedestrian tries to recognize once every 0.3 seconds, the first half of the license plate is tried once every about 0.5 seconds, and the lane departure recognition is tried every about 0.1 second.
The storage period of still images is about 10 seconds. Each cycle can be controlled by the driver, and there is a restriction that the cycle can be changed within the computation capacity of the smart camera of the vehicle. The measurement periods of the image metadata and the sensor data are shown in Table 1.
In addition, when generating image metadata, new information is derived using the image recognition result and included in the metadata.
Typically, the distance from the front vehicle can be estimated based on the result of the license plate recognition. Although the estimated distance is not accurate, it is possible to utilize the increase / decrease ratio of the distance to provide another information and service.
The traffic information collection system based on real-time target detection according to the present invention comprises a smart camera for a vehicle and a traffic information providing server for real-time traffic information service using the same.
The smart camera for a vehicle has a sensor unit for collecting traffic information and a communication unit for sharing the collected data.
The sensor unit can acquire the sensor information of the inside of the vehicle through OBD and acquire the forward image through the camera and analyze it. The communication unit has a function of sharing collected traffic information.
In order to collect traffic information, the sensor recognizes the speed of the vehicle, the distance from the leading vehicle, and the number of vehicles ahead. The traffic information of the current road is generated by combining the collected information, and is shared with other vehicles through the communication unit.
And OBD is used to read speed value to measure vehicle speed. OBD is a system that is used to diagnose faults of a vehicle. It is a system that can read current sensor values in a vehicle.
In order to measure the distance from the leading vehicle, the smart camera for the vehicle analyzes the image collected from the camera to recognize the license plate of the preceding vehicle.
Estimates the distance from the front vehicle based on the position of the license plate recognized in the input image.
The formula for converting the license plate position to the distance from the front vehicle is as follows.
In the above equation, F dist is the distance from the front vehicle, d pixel is the distance between the end of the image and the license plate, and α and β are constants.
The method of recognizing the number of vehicles in front is estimated by the number of license plate recognized in the image.
The traffic information providing server generates the traffic volume information of the road based on the three pieces of information collected through the above process.
The formula for generating the traffic volume information based on the collected traffic information is as follows.
In the above equation, C traffic is the traffic volume information, w s , w n , and w d are the weights of the vehicle speed, the number of vehicles in the front, and the distance information items in the vehicle.
S (x), N (y), and D (z) are functions for scoring the vehicle speed, the number of vehicles ahead, and the distance information between vehicles, respectively.
The scoring function is selected considering the effect of each traffic information on road traffic volume.
If the vehicle speed is above a certain level, the driver's tendency is more affected than the degree of congestion of the road. Therefore, it should be scored sensitively in the low and middle sections.
As the number of vehicles ahead increases beyond the acceptable range of the road, the degree of stagnation increases sharply and must be scored sensitively in the middle section in the high section.
Finally, the distance information between vehicles should be sensitively scored in the middle section because the effect on road congestion is low at or above a certain distance.
The scoring function selected in consideration of the characteristics of this information is as follows.
The traffic information processing configuration and method of the traffic information collection system by real-time target detection according to the present invention will be described in detail as follows.
FIG. 6 is a block diagram of a traffic information collection system based on real-time target detection according to the present invention, and FIG. 7 is a flowchart illustrating a method of collecting traffic information by real-time target detection according to the present invention.
As shown in FIG. 6, the traffic information processing system of the traffic information collection system using real-time target detection according to the present invention includes a
As shown in FIG. 7, in the traffic information collection method using real-time target detection according to the present invention, when an image (frame) is input (S701), it is checked whether a frame received by the dynamic operation frame adjustment technique is an object to be inspected. S702)
For example, image processing can be performed every 3 frames by a dynamic operation frame adjustment technique.
In order to distinguish the fixed region image processing or the dynamic region image processing, it is determined whether the region is a dynamic region (S703)
If the object is found in the previous frame and the dynamic region processing is not performed consecutively N times, it is judged to be a dynamic region.
If it is determined that the area is not the dynamic area, the fixed area image processing is performed (S704)
Fixed area image processing recognizes pedestrians, front license plates, front signs, lanes, etc., and sets the fixed area of interest and location to fit the characteristics of each object.
Pedestrian recognition recognizes from horizontal (2/6 to 5/6) and vertical (4/8 to 7/8) based on the input image.
The license plate recognition recognizes horizontal (2/6 to 5/6) and vertical (4/8 to 7/8) based on the input image.
And the sign recognition recognizes from the input image as horizontal (1/2 point to 2/2 point) and vertical (0 point to 1/2 point).
And the lane recognition recognizes from horizontal (3/20 point to 17/20 point) and vertical (6/8 point to 7/8 point) based on the input image.
If it is determined to be the dynamic region, the dynamic region image processing is performed (S705)
In the dynamic region image processing, the ROI is set considering the movement of the object based on the stored target position, and the ROI is set by increasing the smallest rectangle including the recognized ROI by 30px.
When the object is found by performing the fixed region image processing and the dynamic region image processing in this manner (S706), the position of the object obtained from the image processing result is stored in a rectangular shape (coordinates of the upper left point and coordinates of the lower right point). (S707)
The system and method for collecting traffic information by real-time target detection according to the present invention as described above includes a method of dynamically changing the region of interest to reduce the amount of computation and a real-time target detection .
The present invention can provide a safety and convenience of a user by integrating real-time target detection technology for advanced driver assistance system and vehicle technology related to real-time traffic information collection device and method using the same.
As described above, it will be understood that the present invention is implemented in a modified form without departing from the essential characteristics of the present invention.
It is therefore to be understood that the specified embodiments are to be considered in an illustrative rather than a restrictive sense and that the scope of the invention is indicated by the appended claims rather than by the foregoing description and that all such differences falling within the scope of equivalents thereof are intended to be embraced therein It should be interpreted.
60.
62.
64. Target position storage section
Claims (13)
And a traffic information providing server for analyzing traffic information received from a smart camera for a vehicle to provide a user with a real time traffic information service regarding the photograph of the current road and the degree of congestion,
The traffic information providing server calculates traffic volume information based on the speed of the vehicle, the distance to the head vehicle,
C traffic is traffic volume information, w s , w n , and w d are weights of the vehicle speed, the number of vehicles in front, and the distance information items of vehicles, and the sum of the three weights is 1, S (x) D (z) is a function for scoring the vehicle speed, the number of vehicles ahead, and the distance information between vehicles,
Wherein the vehicle smart camera dynamically changes a region of interest and adjusts a frame to be processed in the input image to perform real time target detection.
The number of frames to be operated (FPF) to select the number of frames to be processed among consecutive frames input based on the vehicle speed and the amount of change in acceleration,
Here, S is the speed of the vehicle, alpha is the fine impact correction coefficient, A is the acceleration change amount, and FR is the frame rate of the target value.
Extracts fixed regions corresponding to the characteristics of each target from the entire input image,
And a target is detected while moving a region of interest according to a change in the position of the target.
Wherein the traffic information is measured based on the speed of the vehicle, the distance from the leading vehicle, and the number of the preceding vehicle when the road traffic is measured.
An AP (Application Processor) for real-time image processing, a coprocessor for controlling the sensors,
A sensor unit including a camera, an acceleration sensor, and a GPS (Global Positioning System) module used to acquire vehicle periphery information;
A mobile communication network module for transmitting collected traffic information to a traffic information providing server, a Wi-Fi module for informing a driver of a dangerous situation through a driver's smart phone, and an OBD scanner The traffic information collection system according to claim 1,
An image processing program for detecting a target object from an image and storing the target object in a database,
A media transmission program for inquiring a list of still images in the database, and transmitting the media files that have not been transmitted to the traffic information providing server,
And a data transmission program for transmission of general data except media transmission is mounted.
By analyzing the surrounding situation in real time through image recognition, the result of image analysis made of image metadata is notified to the driver and stored in the database,
Sensor values including OBD are periodically sampled to generate sensor data, store it in a database,
Wherein the image metadata and the sensor data are combined and transmitted to the traffic information server.
In order to measure the distance from the leading vehicle, the collected images are analyzed to recognize the license plate of the preceding vehicle, and the distance from the preceding vehicle is estimated based on the position of the license plate recognized in the input image,
Translating the license plate position to the distance to the front vehicle
A frame checking unit for checking whether an inputted frame is to be inspected for dynamic operation frame adjustment;
An image processing region judging unit for judging an image processing region to separately perform the fixed region image processing or the dynamic region image processing;
An image processing unit for performing the setting and recognition of a region of interest by classifying the fixed region image processing or the dynamic region image processing;
A target detection unit for detecting a target through setting and recognizing a region of interest;
And a target position storage unit for storing the position of the object obtained as a result of the image processing in a rectangular form,
In order to change the area of interest dynamically, the number of frames (FPF) to be selected and operated among the continuous frames inputted based on the vehicle speed and the amount of change in acceleration is calculated by the following equation
Here, S is the speed of the vehicle, alpha is the fine impact correction coefficient, A is the acceleration change amount, and FR is the frame rate of the target value.
Checking whether an input frame is an inspection target when a video is input;
Determining whether the image is a dynamic region to distinguish the fixed region image processing or the dynamic region image processing;
Performing a fixed area image processing if it is determined that the area is not a dynamic area;
Performing dynamic region image processing if it is determined to be a dynamic region;
And storing the position of the object obtained as a result of the image processing in a rectangular form when the object is found by performing the fixed region image processing and the dynamic region image processing,
In order to change the area of interest dynamically, the number of frames (FPF) to be selected and operated among the continuous frames inputted based on the vehicle speed and the amount of change in acceleration is calculated by the following equation
Here, S is a speed of the vehicle,? Is a fine impact correction coefficient,? A is an acceleration change amount, and FR is a frame rate of a target value.
A traffic information plate, a pedestrian, a front license plate, a front sign, and a lane, and setting an area of interest fixed in a position and a size according to characteristics of each object.
Wherein the region of interest is set in consideration of the movement of the object based on the stored target position and the region of interest is set by extending the smallest rectangle including the recognized object to a predetermined size. How to collect traffic information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020150138639A KR101735557B1 (en) | 2015-10-01 | 2015-10-01 | System and Method for Collecting Traffic Information Using Real time Object Detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020150138639A KR101735557B1 (en) | 2015-10-01 | 2015-10-01 | System and Method for Collecting Traffic Information Using Real time Object Detection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| KR20170039465A KR20170039465A (en) | 2017-04-11 |
| KR101735557B1 true KR101735557B1 (en) | 2017-05-15 |
Family
ID=58580628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| KR1020150138639A KR101735557B1 (en) | 2015-10-01 | 2015-10-01 | System and Method for Collecting Traffic Information Using Real time Object Detection |
Country Status (1)
| Country | Link |
|---|---|
| KR (1) | KR101735557B1 (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108124507B (en) * | 2017-12-28 | 2021-02-09 | 深圳市锐明技术股份有限公司 | Demonstration method and device for intelligent vehicle-mounted ADAS function and terminal equipment |
| CN110610514B (en) * | 2018-06-15 | 2023-09-19 | 株式会社日立制作所 | Method, device and electronic equipment for realizing multi-target tracking |
| CN112997211B (en) * | 2018-11-13 | 2024-04-05 | 索尼半导体解决方案公司 | Data distribution system, sensor device, and server |
| KR102266996B1 (en) | 2019-12-10 | 2021-06-18 | 성균관대학교산학협력단 | Method and apparatus for limiting object detection area in a mobile system equipped with a rotation sensor or a position sensor with an image sensor |
| KR102263392B1 (en) * | 2020-01-22 | 2021-06-10 | 이성진 | Vehicle Information Transmission and Reception System for Efficient Traffic Network Control System |
| CN112378697A (en) * | 2020-11-09 | 2021-02-19 | 刘平亮 | Intelligent coking coal sampling system |
| CN112507844B (en) * | 2020-12-02 | 2022-12-20 | 博云视觉科技(青岛)有限公司 | Traffic jam detection method based on video analysis |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002190013A (en) | 2000-12-21 | 2002-07-05 | Nec Corp | System and method for detecting congestion by image recognition |
| KR101372838B1 (en) * | 2013-02-20 | 2014-03-12 | (주)인펙비전 | Multi-function vehicle photographing system and method |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20120075856A (en) | 2010-12-29 | 2012-07-09 | 한국과학기술원 | Intelligent eco-driving apparatus through driving situation awareness and method therefor |
| KR20150043934A (en) | 2013-10-15 | 2015-04-23 | 삼성전자주식회사 | Method and apparatus for control brightness of display |
-
2015
- 2015-10-01 KR KR1020150138639A patent/KR101735557B1/en active IP Right Grant
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002190013A (en) | 2000-12-21 | 2002-07-05 | Nec Corp | System and method for detecting congestion by image recognition |
| KR101372838B1 (en) * | 2013-02-20 | 2014-03-12 | (주)인펙비전 | Multi-function vehicle photographing system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| KR20170039465A (en) | 2017-04-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR101735557B1 (en) | System and Method for Collecting Traffic Information Using Real time Object Detection | |
| CN111554088B (en) | Multifunctional V2X intelligent roadside base station system | |
| US11315420B2 (en) | Moving object and driving support system for moving object | |
| KR101925293B1 (en) | The vehicle detecting system by converging radar and image | |
| KR101446546B1 (en) | Display system of vehicle information based on the position | |
| US9207088B2 (en) | Method for operating a motor vehicle and motor vehicle | |
| CN112562405A (en) | Radar video intelligent fusion and early warning method and system | |
| CN111754581A (en) | Camera calibration method, roadside sensing equipment and intelligent traffic system | |
| CN108645375B (en) | Rapid vehicle distance measurement optimization method for vehicle-mounted binocular system | |
| KR101898051B1 (en) | Multilane vehicle speed detecting system | |
| US20210003419A1 (en) | System for Generating Confidence Values in Digital Road Maps | |
| KR20140144047A (en) | System and method for estimating traffic characteristics information in the road network on a real-time basis | |
| CN114359181A (en) | Intelligent traffic target fusion detection method and system based on image and point cloud | |
| KR20190043396A (en) | Method and system for generating and providing road weather information by using image data of roads | |
| CN111319560B (en) | Information processing system, program, and information processing method | |
| US11837084B2 (en) | Traffic flow estimation apparatus, traffic flow estimation method, traffic flow estimation program, and storage medium storing traffic flow estimation program | |
| Zhuang et al. | Edge-based traffic flow data collection method using onboard monocular camera | |
| KR102076316B1 (en) | Road Surface Condition Management System | |
| CN113177976A (en) | Depth estimation method and device, electronic equipment and storage medium | |
| KR102018582B1 (en) | The apparatus and method for each lane collecting traffic information | |
| CN112241004B (en) | Object recognition device | |
| JP2022027305A (en) | Risky driving detection device, risky driving detection system, and risky driving detection program | |
| JP2022056153A (en) | Temporary stop detection device, temporary stop detection system, and temporary stop detection program | |
| CN109426771A (en) | The device and method that the wisp region of vehicle periphery is identified | |
| JP2012216125A (en) | Object detection method, object detection device, and object detection program |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| A201 | Request for examination | ||
| E902 | Notification of reason for refusal | ||
| E701 | Decision to grant or registration of patent right | ||
| GRNT | Written decision to grant |