KR-20260063922-A - INTELLIGENT VEHICLE RECOGNITION SYSTEM APPLYING TO INTELLIGENT TRANSPORT SYSTEM AND THEREOF

KR20260063922AKR 20260063922 AKR20260063922 AKR 20260063922AKR-20260063922-A

Abstract

Intelligent vehicle recognition technology, particularly intelligent vehicle recognition technology applied to intelligent transportation systems, generates multiple tracks of vehicles traveling on the road based on a point cloud created by 4D radar signal processing, analyzes the distribution of the tracks, and determines the type of vehicle traveling from the distribution characteristics of the tracks.

Inventors

김근환
김용재

Assignees

(주)스마트레이더시스템

Dates

Publication Date: 20260507
Application Date: 20241031

Claims (16)

In an intelligent vehicle recognition device applied to an Intelligent Transport System, An antenna assembly comprising: a transmitting antenna group including a plurality of transmitting antennas that transmit a radar signal toward a vehicle moving in a lane direction; and a receiving antenna group including a plurality of receiving antennas that receive a radar signal reflected from the vehicle; A radar signal processing circuit that processes radar signals received from a receiving antenna group and outputs a radar point cloud containing velocity information and location information of each point; A track generation and detection circuit that generates and detects multiple tracks representing at least a portion of a vehicle by clustering a group of densely packed radar point clouds having similar characteristics in the direction of a vehicle lane within a designated vehicle recognition section; and A vehicle type determination circuit that determines the type of vehicle by recognizing the approximate outline of the vehicle based on the distribution of multiple tracks in the horizontal and vertical directions; An intelligent vehicle recognition device including
In claim 1, the antenna assembly is: An intelligent vehicle recognition device comprising a plurality of transmitting antennas or a plurality of receiving antennas spaced apart in the horizontal and vertical directions to receive a radar signal including information on the angle of arrival in the azimuth direction and the elevation direction.
In claim 1, the track forming and detection circuit is: Track accumulation circuit that accumulates tracks included in multiple frames; An intelligent vehicle recognition device including
In claim 1, the vehicle type determination circuit is: An intelligent vehicle recognition device comprising: a track feature classification circuit that determines the type of vehicle based on at least one of the width, height, distance between center points of consecutive tracks, and the number of tracks of each track belonging to the vehicle.
In claim 1, the vehicle type determination circuit is: A first large vehicle classification circuit that, when the determined vehicle is a large vehicle, determines the type of large vehicle by classifying a 2D image obtained by projecting the spatial distribution of accumulated tracks of the recognized vehicle in three directions (front, top, and side) using a machine learning algorithm; and A second large vehicle classification circuit that determines the type of a large vehicle based on machine learning, wherein, when the determined vehicle is a large vehicle, it further determines the horizontal and vertical distribution of tracks formed and detected in an extended recognition section, characterized by at least one of the width, height, distance between center points of consecutive tracks, and number of tracks of each track belonging to the vehicle; An intelligent vehicle recognition device comprising at least one more of the following.
In claim 1, the intelligent vehicle recognition system is: A valid track selection circuit that removes at least one of a ghost track, a noise track, and a track not containing a license plate among a plurality of tracks, and selects a track containing a license plate as a valid track; An intelligent vehicle recognition device further comprising
In claim 3, the intelligent vehicle recognition system is: A vehicle speed calculation circuit that calculates the vehicle speed from the positions of identical tracks existing in consecutive frames and the Doppler values of those tracks; An intelligent vehicle recognition device further comprising
In claim 7, the intelligent vehicle recognition system is: An intelligent vehicle recognition device further comprising: a license plate position output circuit that outputs the position and speed of a license plate and the type of vehicle from interest track information determined by an interest track determination circuit when the vehicle speed calculated by the vehicle speed calculation circuit exceeds a reference speed.
In an intelligent vehicle recognition method (S1000) applied to an intelligent transport system, A radar transmission and reception step (S100) in which a radar signal is transmitted toward a vehicle moving in a lane direction by a transmitting antenna group including a plurality of transmitting antennas, and a radar signal reflected from the vehicle is received by a receiving antenna group including a plurality of receiving antennas; A radar signal processing step (S200) that processes radar signals received from a receiving antenna group and outputs a radar point cloud containing velocity information and location information of each point; A track formation and detection step (S300) in which a group of densely clustered radar point clouds having similar characteristics in the direction of the vehicle lane in a designated vehicle recognition section is clustered to generate and detect a plurality of tracks representing at least a part of the vehicle; and A vehicle type determination step (S500) in which the approximate outline of a vehicle is recognized and the type of vehicle is determined based on the distribution of multiple tracks in the horizontal and vertical directions; An intelligent vehicle recognition method including
In claim 9, the radar transmission and reception step (S100) is: An intelligent vehicle recognition method in which a signal capable of measuring the angle of arrival in the azimuth direction and the elevation direction is received from a receiving antenna at a plurality of transmitting antennas or a plurality of receiving antennas arranged spaced apart in the horizontal and vertical directions.
In claim 9, the track formation and detection step (S300) is: Track accumulation step (S310), in which tracks included in multiple frames are accumulated; An intelligent vehicle recognition method including
In claim 9, the vehicle type determination step (S500) is: A track feature classification step (S510) in which the type of vehicle is determined based on at least one of the width, height, distance between center points of consecutive tracks, and number of tracks of each track belonging to the vehicle; An intelligent vehicle recognition method including
In claim 9, the vehicle type determination step (S500) is: A first large vehicle classification step (S520-1), wherein if the determined vehicle is a large vehicle, the type of large vehicle is determined by classifying the 2D image obtained by projecting the spatial distribution of the accumulated tracks of the recognized vehicle in three directions (front, top, and side) using a machine learning algorithm; and A second large vehicle classification step (S520-2) for determining the type of large vehicle based on machine learning, wherein if the determined vehicle is a large vehicle, the horizontal and vertical distribution of tracks formed and detected in an extended recognition section is further based on the width, height, distance between center points of consecutive tracks, and number of tracks of each track belonging to the vehicle. An intelligent vehicle recognition method comprising at least one more of the following.
In claim 9, the intelligent vehicle recognition method (S1000) is: A valid track selection step (S400) of removing at least one of a ghost track, a noise track, and a track not containing a license plate among a plurality of tracks, and selecting a track containing a license plate as a valid track; An intelligent vehicle recognition method that further includes
In claim 11, the intelligent vehicle recognition method (S1000) is: A vehicle speed calculation step (S600) in which the vehicle speed is calculated from the positions of identical tracks existing in consecutive frames and the Doppler values of those tracks; An intelligent vehicle recognition method that further includes
In claim 15, the intelligent vehicle recognition method (S1000) is: A license plate position output step (S700) in which, if the speed of the vehicle calculated by the vehicle speed calculation circuit exceeds the reference speed, the position and speed of the license plate and the type of the vehicle are output from the valid track information determined by the valid track determination circuit; An intelligent vehicle recognition method that further includes

Description

Intelligent Vehicle Recognition System and Method Applied to Intelligent Transport System The proposed invention discloses intelligent vehicle recognition technology, particularly intelligent vehicle recognition technology applied to intelligent transportation systems. In order to implement a rapid, safe, and comfortable next-generation transportation system in line with the accelerating information society, an Intelligent Transport System (ITS) is applied to analyze traffic conditions in real time using computers. Based on this, road traffic management and the implementation of an optimal signal system are sought, while simultaneously automating tasks such as measuring travel time, identifying traffic accidents, and cracking down on overloading. Among the technologies for enforcing vehicle speed limits, the radar method, which utilizes radar technology to perform speed enforcement, measures vehicle speed by emitting radio waves and analyzing the signals reflected back from the vehicle. Specifically, it calculates the vehicle speed through phase changes of radar waves using the Doppler effect. The radar method is widely used because it offers a short time for acquiring speed information and accurate speed measurement. However, conventional radar systems do not determine the type of vehicle being measured, which presents difficulties in imposing differential fines based on vehicle type. Furthermore, for effective speed enforcement, it is necessary to identify the location of the license plate within the point cloud generated by the radar. Consequently, it is essential to determine whether the vehicle being measured is a large vehicle, a small vehicle, or a motorcycle; thus, there is an urgent need to develop camera radar technology for speed enforcement that can determine the vehicle type. Korean Published Patent Application (Publication No.: 10-2022-0153836, “Traffic Safety Signal Light and Traffic Information Collection System Equipped with AI”) discloses a traffic safety signal light and a traffic information collection system equipped with AI that receives a driving vehicle recognition signal and driving speed detection information from a Doppler radar to recognize pedestrians and driving vehicles and guides them to drive and walk safely, but does not disclose a technology for determining the type of driving vehicle. FIG. 1 illustrates an intelligent vehicle recognition device that transmits and receives radar signals toward a vehicle moving in a lane direction, according to one embodiment. FIG. 2 shows a plurality of tracks generated and detected in a vehicle recognition section by an intelligent vehicle recognition device according to one embodiment. FIG. 3 is a block diagram showing the configuration of an intelligent vehicle recognition device according to one embodiment. FIG. 4 shows an array of transmitting antennas and receiving antennas according to one embodiment. FIG. 5 is a flowchart illustrating an intelligent vehicle recognition method according to one embodiment. The foregoing and additional aspects are embodied through embodiments described with reference to the attached drawings. It is understood that the components of each embodiment may be combined in various ways within the embodiment or with components of other embodiments, unless otherwise stated or contradicted. Based on the principle that the inventor may appropriately define the concepts of terms to best describe his invention, the terms used in this specification and claims shall be interpreted in a meaning and concept consistent with the description or proposed technical idea. Blocks referred to as 'circuits' in this specification may be composed of hardware, such as dedicated semiconductors, gate arrays, or FPGAs, or parts thereof. One or more blocks may be implemented as a single piece of hardware. As another example, these blocks may be implemented in software as an information processing device in which a computation element executes program instructions stored in memory elements. Multiple blocks may be implemented as part of a program executed on the same computation element. As yet another example, these blocks may be implemented in a hybrid form in which part of the individual circuit is hardware and part is software. Furthermore, in the software implementation, the computation element may include digital signal processors, dedicated computation processors, artificial intelligence processing engines, dedicated artificial intelligence processors, graphics processors, or a combination thereof to the extent possible. Preferred embodiments of the present invention will be described in detail below with reference to the attached drawings. FIG. 1 illustrates an intelligent vehicle recognition device that transmits and receives radar signals toward a vehicle moving in a lane direction according to one embodiment. As illustrated, the intelligent vehicle recognition device (1000) can transmit a radar signal by irradiating a radar