Search

CN-121999618-A - Vehicle gesture recognition method and device

CN121999618ACN 121999618 ACN121999618 ACN 121999618ACN-121999618-A

Abstract

The embodiment of the invention discloses a vehicle posture identification method and device, which are characterized in that a vehicle returning image is acquired, a parking line is identified on the vehicle returning image based on a predetermined vehicle returning verification mode, parking posture information is determined to carry out vehicle returning verification according to the parking posture information, wherein the vehicle returning verification mode is determined according to the wearing degree of the parking line of an area where a vehicle is located, and the parking posture information comprises the distance between a vehicle body and the parking line and the included angle between the vehicle body direction and the parking line. Therefore, according to the embodiment, the parking line recognition is carried out by determining different parking line abrasion degrees of all areas and checking modes, the accuracy of vehicle gesture recognition is improved, and the user experience is improved while normalized parking is realized.

Inventors

  • SONG YUN
  • LIU JINJING

Assignees

  • 北京嘀嘀无限科技发展有限公司

Dates

Publication Date
20260508
Application Date
20260211

Claims (15)

  1. 1. A vehicle posture recognition method, characterized in that the method comprises: Acquiring a vehicle returning image; Carrying out parking line identification on the vehicle returning image based on a predetermined vehicle returning checking mode, wherein the vehicle returning checking mode is determined according to the abrasion degree of a parking frame line of the area where the vehicle is located; and determining parking posture information, wherein the parking posture information comprises the distance between a vehicle body and a parking line and the included angle between the direction of the vehicle body and the parking line.
  2. 2. The method of claim 1, wherein said identifying a stop line for said return vehicle image comprises: Identifying the returning image based on a pre-trained image processing model, and acquiring at least one identification frame; And performing center line fitting on each identification frame to obtain the parking line.
  3. 3. The method of claim 2, wherein said performing a centerline fit on each of said identification frames to obtain said parking line comprises: and in response to the existence of two or more identification frames, performing linear fitting on the center point coordinates of each identification frame to acquire the parking line.
  4. 4. The method of claim 2, wherein said performing a centerline fit on each of said identification frames to obtain said parking line comprises: responsive to the presence of an identification frame, acquiring a center point of the identification frame and a selection point located within the identification frame on at least one side of the center point; and carrying out linear fitting on the central point and each selected point to obtain the parking line.
  5. 5. The method of claim 2, wherein the image processing model is an image detection model, the identification frame is a detection frame determined by the image detection model, or the image processing model is an image segmentation model, the identification frame is a segmentation area determined by the image segmentation model, or the image processing model comprises an image detection model and an image segmentation model, and the identification frame comprises a detection frame determined by the image detection model and a segmentation area determined by the image segmentation model.
  6. 6. The method of claim 1, wherein the determining parking pose information comprises: Calculating the vertical distance from a vehicle body calibration point to the parking line, wherein the vehicle body calibration point is determined based on the zero point calibrated by a corresponding camera; The vertical distance in the image coordinate system is mapped to the actual distance.
  7. 7. The method of claim 1, wherein the determining parking pose information comprises: and determining an included angle between the vertical direction of the image coordinate system and the parking line as an included angle between the vehicle body direction and the parking line.
  8. 8. The method of claim 6, wherein determining a distance of the vehicle body from the stop line further comprises: And compensating the actual distance based on a predetermined vehicle type distance compensation parameter to acquire the distance between the vehicle body and the parking line.
  9. 9. The method according to claim 1, wherein the method further comprises: and responding to failure in identifying the vehicle posture information corresponding to the current vehicle returning image, wherein the image identification times are smaller than a preset value, and displaying or broadcasting prompt information which is used for prompting a user to acquire the vehicle returning image again after correcting the vehicle posture.
  10. 10. The method of claim 1, wherein the return checking mode comprises a device side checking mode, a cloud checking mode, a serial checking mode of a device side and a cloud, and a parallel checking mode of the device side and the cloud; The vehicle gesture recognition method is executed by the equipment end through the equipment end verification mode representation, the vehicle gesture recognition method is executed by the cloud end through the cloud end verification mode representation, when the equipment end does not successfully recognize the vehicle gesture information through the serial verification mode representation of the equipment end and the cloud end, the cloud end carries out the vehicle gesture recognition method based on the vehicle returning image reported by the equipment end, and the parallel verification mode representation equipment end and the cloud end synchronously execute the vehicle gesture recognition method.
  11. 11. The method according to claim 10, wherein the device side is deployed with a pre-trained image detection model, and the cloud is deployed with an image segmentation model, or with an image segmentation model and an image detection model.
  12. 12. A vehicle posture identifying apparatus, characterized in that the apparatus comprises: an image acquisition unit configured to acquire a return image; The identification unit is configured to identify a parking line of the vehicle returning image based on a predetermined vehicle returning verification mode, wherein the vehicle returning verification mode is determined according to the abrasion degree of the parking line of the area where the vehicle is located; And a posture determining unit configured to determine parking posture information including a distance of a vehicle body from a parking line and an angle of a vehicle body direction from the parking line.
  13. 13. An electronic device comprising a memory and a processor, wherein the memory is configured to store one or more computer program instructions, wherein the one or more computer program instructions are executed by the processor to implement the method of any of claims 1-11.
  14. 14. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when executed by a processor, implements the method according to any of claims 1-11.
  15. 15. A computer program product, characterized in that the computer program product, when run on a computer, causes the computer to perform the method according to any of claims 1-11.

Description

Vehicle gesture recognition method and device Technical Field The invention relates to the technical field of intelligent transportation, in particular to a vehicle gesture recognition method and device. Background With the popularization of shared vehicles (e.g., two-wheeled vehicles, etc.), the management of the specifications of a large number of shared vehicles has become a non-negligible problem. At present, parking management of shared vehicles is realized by setting a parking area in a city and checking that the vehicles are parked in the parking area only when the vehicles are permitted to park, but due to the problems of positioning accuracy or gesture recognition accuracy, the problems that the vehicles are stopped in the parking area normally but the checking is not passed and the like exist, and bad experience is brought to users. Disclosure of Invention In view of the above, embodiments of the present invention provide a vehicle gesture recognition method and apparatus, so as to determine different parking line verification modes according to the wear degree of the parking line in each area to perform parking line recognition, thereby improving accuracy of vehicle gesture recognition, and improving user experience while realizing normalized parking. In a first aspect, an embodiment of the present invention provides a vehicle gesture recognition method, including: Acquiring a vehicle returning image; Carrying out parking line identification on the vehicle returning image based on a predetermined vehicle returning checking mode, wherein the vehicle returning checking mode is determined according to the abrasion degree of a parking frame line of the area where the vehicle is located; and determining parking posture information, wherein the parking posture information comprises the distance between a vehicle body and a parking line and the included angle between the direction of the vehicle body and the parking line. In a second aspect, an embodiment of the present invention provides a vehicle gesture recognition apparatus, the apparatus including: an image acquisition unit configured to acquire a return image; The identification unit is configured to identify a parking line of the vehicle returning image based on a predetermined vehicle returning verification mode, wherein the vehicle returning verification mode is determined according to the abrasion degree of the parking line of the area where the vehicle is located; And a posture determining unit configured to determine parking posture information including a distance of a vehicle body from a parking line and an angle of a vehicle body direction from the parking line. In a third aspect, embodiments of the present invention provide an electronic device comprising a memory for storing one or more computer program instructions, and a processor, wherein the one or more computer program instructions are executed by the processor to implement a method as described above. In a fourth aspect, embodiments of the present invention provide a computer readable storage medium having stored therein a computer program which when executed by a processor implements a method as described above. In a fifth aspect, embodiments of the present invention provide a computer program product which, when run on a computer, causes the computer to perform the method as described above. According to the embodiment, a vehicle returning image is acquired, a parking line is identified on the vehicle returning image based on a predetermined vehicle returning checking mode, the distance between a vehicle body and the parking line and the included angle between the vehicle body direction and the parking line are determined, and the vehicle posture is determined to accord with the vehicle returning condition in response to the fact that the distance and the included angle meet the predetermined condition, wherein the vehicle returning checking mode is determined according to the abrasion degree of the parking line in the area where the vehicle is located. Therefore, according to the embodiment, the parking line recognition is carried out by determining different parking line abrasion degrees of all areas and checking modes, the accuracy of vehicle gesture recognition is improved, and the user experience is improved while normalized parking is realized. Drawings The above and other objects, features and advantages of the present invention will become more apparent from the following description of embodiments of the present invention with reference to the accompanying drawings, in which: FIG. 1 is a flow chart of a vehicle gesture recognition method of an embodiment of the present invention; FIG. 2 is a flow chart of a park line identification method according to an embodiment of the invention; FIG. 3 is a schematic illustration of a park line confirmation process according to an embodiment of the invention; FIG. 4 is a schematic diagram of another park line confirmation process ac