Search

CN-121982907-A - Intelligent parking system and vehicle parking management method

CN121982907ACN 121982907 ACN121982907 ACN 121982907ACN-121982907-A

Abstract

The invention discloses an intelligent parking system and a vehicle parking management method, which relate to the technical field of intelligent transportation and Internet of things, and comprise the following steps that S1, vehicle position signals and vehicle image characteristics are collected and are subjected to data fusion to obtain real-time coordinates of a vehicle with direction information; S2, constructing a continuous track chain based on real-time coordinates of the vehicle, and identifying abnormal behavior of the vehicle based on a preset track characteristic threshold, and S3, generating and executing a corresponding barrier gate control instruction according to the abnormal behavior of the vehicle. By fusing geomagnetic and video sensing data in a dynamic self-adaptive mode, a vehicle continuous track chain is constructed, abnormal behaviors such as vehicle detention, turning around and the like can be accurately identified, real-time linkage with barrier gate control is realized, and therefore safety of vehicle smashing prevention, accuracy of illegal behavior identification and real-time performance of system response are effectively improved.

Inventors

  • SUN LINGFEI

Assignees

  • 深圳市大道至简信息技术有限公司

Dates

Publication Date
20260505
Application Date
20260316

Claims (10)

  1. 1. An intelligent parking system and a vehicle parking management method are characterized by comprising the following steps: S1, collecting a vehicle position coordinate sequence and an image coordinate sequence, carrying out space-time alignment on the coordinate sequence, dynamically calculating fusion weights according to the fluctuation degree of the vehicle position coordinate sequence and the recognition confidence coefficient of the image, and carrying out weighted fusion on the aligned coordinates based on the fusion weights to generate real-time coordinates of the vehicle with direction information; S2, constructing a track chain composed of continuous track points according to a real-time coordinate of the vehicle in time sequence, and identifying whether abnormal behavior of the vehicle exists or not by analyzing whether the position change and the direction angle change of the track points in the track chain meet a preset track characteristic threshold value or not; and S3, generating and executing a corresponding barrier gate control instruction according to the abnormal behavior of the vehicle.
  2. 2. The intelligent parking system and the vehicle parking management method according to claim 1, wherein S1 comprises a data acquisition and processing step, a fusion weight calculating step and a data fusion and direction calculating step.
  3. 3. The intelligent parking system and vehicle parking management method as set forth in claim 2, wherein the data collecting and processing step includes: Collecting a vehicle position coordinate sequence according to a preset first time interval by using a geomagnetic detector, and collecting an image coordinate sequence containing vehicle feature points according to a preset second time interval by using a video pile; aligning the vehicle position coordinate sequence with the image coordinate sequence based on the time stamp, and converting the vehicle position coordinate sequence and the image coordinate sequence into a unified preset parking lot terrain coordinate system; And determining the fluctuation degree of geomagnetic signals by calculating coordinate difference values of continuously collected data in the vehicle position coordinate sequence, and determining the recognition confidence coefficient of the vehicle feature points in the image coordinate sequence based on the recognition result of the vehicle detection model.
  4. 4. The smart parking system and vehicle parking management method according to claim 3, wherein the fusion weight calculating step includes: converting the geomagnetic quality score into a geomagnetic quality score according to a preset rule and the fluctuation degree of geomagnetic signals; taking the confidence of video identification as a video quality score; Calculating the geometric average value of geomagnetic quality scores and video quality scores to obtain a credibility factor used for representing the current data fusion quality; And comparing the reliability factor with a preset threshold value, and dynamically selecting a fusion weight distribution scheme taking geomagnetic data as a dominant one, video data as a dominant one or balanced one according to a comparison result.
  5. 5. The smart parking system and vehicle parking management method as claimed in claim 4, wherein the data fusion and direction calculation step includes: Based on the fusion weight determined by the fusion weight distribution scheme, carrying out fusion calculation on the aligned vehicle position coordinates and the image coordinates from the same moment to generate vehicle real-time coordinates; and determining the real-time running direction angle of the vehicle according to the change of the real-time coordinates of a plurality of continuous vehicles.
  6. 6. The intelligent parking system and vehicle parking management method as claimed in claim 1, wherein the abnormal behavior of the vehicle includes at least vehicle retention and entrance turning around; the S2 comprises a track chain construction step, an abnormal behavior judgment step and a recognition result generation step.
  7. 7. The smart parking system and vehicle parking management method as claimed in claim 6, wherein the track chain construction step includes: sorting according to the time stamps based on real-time coordinates of the vehicles to form a coordinate sequence arranged in time sequence; And extracting a continuous track point set from the coordinate sequence based on a preset sampling interval or a preset number of coordinate points to form a track chain representing a continuous moving path of the vehicle.
  8. 8. The smart parking system and the vehicle parking management method according to claim 7, wherein the abnormal behavior determination step includes; judging whether the vehicle is in a preset road brake control area or not based on the track chain; When the vehicle is in the road brake control area, determining whether the vehicle has a detention behavior by analyzing the position change and the direction angle change of a plurality of continuous track points in the track chain; When the vehicle is in the entrance passage area, whether the vehicle has turning behavior or not is judged by analyzing the change trend of the running direction angle and the coordinate moving direction of the track chain in a preset time window.
  9. 9. The smart parking system and the vehicle parking management method as claimed in claim 8, wherein the recognition result generating step includes: And according to the judging result of the abnormal behavior judging step, packaging the abnormal behavior type, the time window information according to the judging basis, the triggered threshold condition, the time stamp of occurrence of the abnormality and the associated vehicle identification information into a structured vehicle abnormal behavior event record.
  10. 10. The smart parking system and the vehicle parking management method according to claim 1, wherein S3 includes: generating a barrier gate control instruction according to a preset corresponding control strategy based on the identification result and type of the abnormal behavior of the vehicle; and sending the barrier control instruction to a barrier controller to execute corresponding barrier opening and closing or state maintaining operation.

Description

Intelligent parking system and vehicle parking management method Technical Field The invention relates to the technical field of intelligent transportation and Internet of things, in particular to an intelligent parking system and a vehicle parking management method. Background With the acceleration of the urban process and the continuous increase of the conservation quantity of motor vehicles, an unattended parking lot becomes a mainstream form of intelligent parking. At present, the industry generally adopts a scheme of combining a geomagnetic detector with sensing equipment such as a video pile and the like so as to realize basic functions such as vehicle detection, license plate recognition, automatic charging, barrier gate linkage and the like. Such systems typically operate based on a "single perception triggered simple control" mode, for example, at the entrance, the system is released by lifting the rod after identifying the license plate via video, at the exit, the rod is triggered after charging is completed, and the protection against crashing the vehicle is mainly dependent on a single infrared sensor carried by the barrier gate body. However, such prior art solutions still have significant limitations. For example, the prior art scheme only stays at the basic parking space occupation detection and license plate recognition level, a closed-loop management and control technical system aiming at the vehicle track is not formed, and the prior art scheme still has a technical short board in core scenes such as anti-smashing vehicle safety protection, entrance illegal turning recognition and the like because the perception data are isolated and control logic is extensive, so that the management and control efficiency and the safety reliability are difficult to balance, the prior art is mostly the conventional integration of mature components, the technical innovation design aiming at the pertinence is lacking, and the actual requirements of high-safety unattended scenes cannot be met. Therefore, the reliability, accuracy and instantaneity of the prior art scheme are required to be improved when the technical scheme is applied to an unattended scene with high safety and high order requirements. Disclosure of Invention Based on the above-mentioned drawbacks of the prior art, an objective of the present invention is to provide an intelligent parking system and a vehicle parking management method for solving the above-mentioned technical problems. In order to achieve the above purpose, the invention provides a smart parking system and a vehicle parking management method, comprising the following steps: S1, collecting a vehicle position coordinate sequence and an image coordinate sequence, carrying out space-time alignment on the coordinate sequence, dynamically calculating fusion weights according to the fluctuation degree of the vehicle position coordinate sequence and the recognition confidence coefficient of the image, and carrying out weighted fusion on the aligned coordinates based on the fusion weights to generate real-time coordinates of the vehicle with direction information; S2, constructing a track chain composed of continuous track points according to a real-time coordinate of the vehicle in time sequence, and identifying whether abnormal behavior of the vehicle exists or not by analyzing whether the position change and the direction angle change of the track points in the track chain meet a preset track characteristic threshold value or not; and S3, generating and executing a corresponding barrier gate control instruction according to the abnormal behavior of the vehicle. The invention further provides that the S1 comprises a data acquisition and processing step, a fusion weight calculating step and a data fusion and direction calculating step. The invention is further arranged that the data acquisition and processing steps comprise: Collecting a vehicle position coordinate sequence according to a preset first time interval by using a geomagnetic detector, and collecting an image coordinate sequence containing vehicle feature points according to a preset second time interval by using a video pile; aligning the vehicle position coordinate sequence with the image coordinate sequence based on the time stamp, and converting the vehicle position coordinate sequence and the image coordinate sequence into a unified preset parking lot terrain coordinate system; the fluctuation degree of geomagnetic signals is determined by calculating coordinate differences of continuously collected data in a vehicle position coordinate sequence, and the recognition confidence coefficient of the vehicle feature points in an image coordinate sequence is determined based on recognition results of the vehicle detection model. The invention further provides that the fusion weight calculation step comprises the following steps: converting the geomagnetic quality score into a geomagnetic quality score according to a preset