CN116311144A - Method and device for predicting vehicle steering and computer readable storage medium - Google Patents

Method and device for predicting vehicle steering and computer readable storage medium Download PDF

Info

Publication number
CN116311144A
CN116311144A CN202211641437.4A CN202211641437A CN116311144A CN 116311144 A CN116311144 A CN 116311144A CN 202211641437 A CN202211641437 A CN 202211641437A CN 116311144 A CN116311144 A CN 116311144A
Authority
CN
China
Prior art keywords
target
vehicle
rectangular frame
real
image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211641437.4A
Other languages
Chinese (zh)
Inventor
于占海
曹斌
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Neusoft Reach Automotive Technology Shanghai Co Ltd
Original Assignee
Neusoft Reach Automotive Technology Shanghai Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Neusoft Reach Automotive Technology Shanghai Co Ltd filed Critical Neusoft Reach Automotive Technology Shanghai Co Ltd
Priority to CN202211641437.4A priority Critical patent/CN116311144A/en
Publication of CN116311144A publication Critical patent/CN116311144A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/60—Type of objects
    • G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
    • G06V20/625—License plates
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Computing Systems (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Traffic Control Systems (AREA)

Abstract

The application provides a vehicle steering prediction method, a prediction device and a computer readable storage medium. The prediction method comprises the following steps: performing first target detection on a target real-time image sent by a vehicle to obtain position information of a first rectangular frame of the target vehicle in the target real-time image, wherein the target real-time image is an image in front of the vehicle, and the first rectangular frame is a minimum rectangle surrounding the target vehicle; intercepting a target real-time image according to the position information of the first rectangular frame to obtain a target image, and performing second target detection on the target image to obtain a second rectangular frame of a target license plate in the target image, wherein the target license plate is a license plate of a target vehicle, and the second rectangular frame is a minimum rectangle surrounding the target license plate; the steering trend of the target vehicle is predicted at least according to the first rectangular frame and the second rectangular frame, so that the problem that the steering trend of the automatic driving vehicle is predicted more complicated in the prior art is solved.

Description

Method and device for predicting vehicle steering and computer readable storage medium
Technical Field
The present application relates to the field of automatic driving, and more particularly, to a vehicle steering prediction method, a vehicle steering prediction apparatus, and a computer-readable storage medium.
Background
The development of autopilot is now well-established, and various autopilot functions are increasingly being applied to actual vehicles, in particular L2-class functions. However, due to the influence of factors such as current chip calculation force and cost control, in low-level automatic driving, the problems of high specificity, high pertinence and the like are required to be solved through a relatively simple algorithm.
There are two solutions in the prior art:
1) Solving the problem of low-level automatic driving function by using a complex algorithm;
2) Each function in low-level autopilot employs a logically simple method and performs parallel processing.
In the running process of an automatic driving vehicle, the prediction of the steering trend of the vehicle has the problems of complex calculation and larger calculation force. Therefore, there is a need for a method that can predict the steering tendency of an autonomous vehicle more simply.
Disclosure of Invention
The present invention provides a method, a device and a computer readable storage medium for predicting steering of a vehicle, so as to solve the problem of complex prediction of steering trend of an automatic driving vehicle in the prior art.
According to an aspect of an embodiment of the present invention, there is provided a prediction method of vehicle steering, the prediction method being applied in a server, the prediction method including: performing first target detection on a target real-time image sent by a vehicle to obtain position information of a first rectangular frame of the target vehicle in the target real-time image, wherein the target real-time image is an image in front of the vehicle, the target vehicle and the vehicle are in the same lane, the target vehicle is in front of the vehicle, and the first rectangular frame is a smallest rectangle surrounding the target vehicle; intercepting the target real-time image according to the position information of the first rectangular frame to obtain a target image, and performing second target detection on the target image to obtain a second rectangular frame of a target license plate in the target image, wherein the target license plate is a license plate of the target vehicle, and the second rectangular frame is a minimum rectangle surrounding the target license plate; and predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame.
Optionally, predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame includes: determining a target area in the first rectangular frame by adopting a big data analysis method; predicting that the target vehicle does not have a steering tendency in the case where the second rectangular frame is entirely in the target region; predicting that the target vehicle has a steering tendency in a case where the second rectangular frame is not at all in the target region; and under the condition that the second rectangular frame part is positioned in the target area, constructing a target coordinate system by taking the central point of the target area as the coordinate origin, and predicting the steering trend of the target vehicle at least according to the target overlapping distance of the target area and the second rectangular frame under the target coordinate system, wherein the target overlapping distance is the overlapping distance of the target area and the second rectangular frame in the horizontal direction under the target coordinate system.
Optionally, when the second rectangular frame is located in the positive direction of the target coordinate system, predicting the steering trend of the target vehicle at least according to the target overlapping distance between the target area and the second rectangular frame in the target coordinate system includes: determining the horizontal distance of the target area in the positive direction under the target coordinate system to obtain a first distance; calculating the ratio of the target overlapping distance to the first distance to obtain a first target ratio; and if the first target ratio is smaller than or equal to the preset threshold value, predicting that the target vehicle does not have steering trend.
Optionally, in a case that the second rectangular frame is located in a negative direction of the target coordinate system, predicting the steering tendency of the target vehicle at least according to a target overlapping distance of the target area and the second rectangular frame in the target coordinate system includes: determining the absolute value of the horizontal distance of the target area in the negative direction under the target coordinate system to obtain a second distance; calculating the ratio of the target overlapping distance to the second distance to obtain a second target ratio; and if the second target ratio is smaller than or equal to the preset threshold value, predicting that the target vehicle does not have steering trend.
Optionally, after the target vehicle is predicted to have a steering tendency for the first time, the prediction method further includes: receiving a plurality of frames of other real-time images, and calculating the first target ratio corresponding to the frames of other real-time images under the target coordinate system, wherein the acquisition time of the other real-time images is later than that of the target real-time images; and sequencing the corresponding first target ratios in time sequence, and determining that the target vehicle has a tendency to turn towards the negative direction of the target coordinate system under the condition that the first target ratios gradually decrease.
Optionally, after the target vehicle is predicted to have a steering tendency for the first time, the prediction method further includes: receiving a plurality of frames of other real-time images, and calculating the second target ratio corresponding to the frames of other real-time images under the target coordinate system, wherein the acquisition time of the other real-time images is later than that of the target real-time images; and sequencing the corresponding second target ratios according to a time sequence, and determining that the target vehicle has a tendency to turn towards the positive direction of the target coordinate system under the condition that the second target ratios gradually decrease.
Optionally, performing first target detection on a target real-time image sent by a vehicle to obtain position information of a first rectangular frame of the target vehicle in the target real-time image, where the first target detection includes: and carrying out first target detection on the target real-time image sent by the vehicle by adopting a first model to acquire the position information of the first rectangular frame of the target vehicle in the target real-time image, wherein the first model is constructed based on a deep learning model.
Optionally, performing a second target detection on the target image to obtain a second rectangular frame of the target license plate in the target image, including: and carrying out second target detection on the target image by adopting a second model to obtain the second rectangular frame of the target license plate in the target image, wherein the second model is constructed based on a deep learning model.
According to another aspect of the embodiment of the present invention, there is also provided a prediction apparatus for vehicle steering, the prediction apparatus being applied to a server, the prediction apparatus including: the first detection unit is used for carrying out first target detection on a target real-time image sent by a vehicle to obtain position information of a first rectangular frame of the target vehicle in the target real-time image, wherein the target real-time image is an image in front of the vehicle, the target vehicle and the vehicle are in the same lane, the target vehicle is in front of the vehicle, and the first rectangular frame is a smallest rectangle surrounding the target vehicle; the second detection unit is used for intercepting the target real-time image according to the position information of the first rectangular frame to obtain a target image, and carrying out second target detection on the target image to obtain a second rectangular frame of a target license plate in the target image, wherein the target license plate is a license plate of the target vehicle, and the second rectangular frame is a minimum rectangle surrounding the target license plate; and the prediction unit is used for predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame.
According to still another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium including a stored program, wherein the program executes any one of the prediction methods of vehicle steering.
In the method for predicting the steering of the vehicle, firstly, a target real-time image sent by a vehicle of the vehicle is received, and first target detection is carried out on the received target real-time image to obtain the position information of a first rectangular frame surrounding the target vehicle in the target real-time image; then, based on the position information of the first rectangular frame, intercepting and obtaining a target image in the target real-time image, and carrying out second target detection on the target image to obtain the position information of a second rectangle surrounding a target license plate in the target image; and finally, predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame. In the prediction method, only a first rectangular frame surrounding a target vehicle in a target real-time image is required to be obtained, and a second rectangular frame surrounding a target license plate in the target image is required to be obtained, and then the steering trend of the target vehicle is predicted at least according to the first rectangular frame and the second rectangular frame, so that the target license plate based on the target vehicle is realized, and the steering trend of the target vehicle is predicted. The prediction method of the invention realizes the prediction of the steering trend of the target vehicle through simpler calculation logic, thereby solving the problem of more complex prediction of the steering trend of the automatic driving vehicle in the prior art. The steering trend of the target vehicle is predicted through simpler calculation logic, so that the complexity of calculation of the prediction method is low, the required calculation force is low, and the required cost is low.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:
FIG. 1 illustrates a flow chart of a method of predicting vehicle steering in accordance with one embodiment of the present application;
FIG. 2 illustrates a schematic diagram of a positional relationship of a target vehicle and a host vehicle according to one embodiment of the present application;
FIG. 3 illustrates a schematic diagram of predicting a steering tendency of a target vehicle based on a first rectangular box and a second rectangular box in accordance with an embodiment of the present application;
FIG. 4 shows a schematic diagram of a build target coordinate system of an embodiment of the present application;
FIG. 5 shows a schematic view of a second rectangular box of an embodiment of the present application in a positive direction of a target coordinate system;
FIG. 6 shows a schematic diagram of a second rectangular box of an embodiment of the present application in a negative direction of a target coordinate system;
fig. 7 shows a schematic structural diagram of a vehicle steering prediction device according to an embodiment of the present application.
Wherein the above figures include the following reference numerals:
100. a self-propelled vehicle; 101. a target vehicle; 102. a real-time image of the target; 103. a first rectangular frame; 104. a target area; 105. a second rectangular frame; 106. a target coordinate system; 107. a first distance; 108. a target overlap distance; 109. a second distance.
Detailed Description
It should be noted that, in the case of no conflict, the embodiments and features in the embodiments may be combined with each other. The present application will be described in detail below with reference to the accompanying drawings in conjunction with embodiments.
In order to make the present application solution better understood by those skilled in the art, the following description will be made in detail and with reference to the accompanying drawings in the embodiments of the present application, it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by one of ordinary skill in the art based on the embodiments herein without making any inventive effort, shall fall within the scope of the present application.
It should be noted that the terms "first," "second," and the like in the description and claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate in order to describe the embodiments of the present application described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
As described in the background art, in order to solve the above-mentioned problem, in an exemplary embodiment of the present application, a method for predicting steering of a vehicle, a prediction apparatus, and a computer-readable storage medium are provided.
According to an embodiment of the application, a method for predicting vehicle steering is provided.
Fig. 1 is a flowchart of a method of predicting vehicle steering according to an embodiment of the present application. The above prediction method is applied to a server, as shown in fig. 1, and includes the following steps:
step S101, performing first target detection on a target real-time image sent by a vehicle to obtain position information of a first rectangular frame of the target vehicle in the target real-time image, wherein the target real-time image is an image in front of the vehicle, the target vehicle and the vehicle are in the same lane, the target vehicle is in front of the vehicle, and the first rectangular frame is a smallest rectangle surrounding the target vehicle;
step S102, intercepting the target real-time image according to the position information of the first rectangular frame to obtain a target image, and carrying out second target detection on the target image to obtain a second rectangular frame of a target license plate in the target image, wherein the target license plate is the license plate of the target vehicle, and the second rectangular frame is the smallest rectangle surrounding the target license plate;
And step S103, predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame.
In the vehicle steering prediction method, firstly, a target real-time image sent by a vehicle of a host vehicle is received, and a first target detection is carried out on the received target real-time image to obtain position information of a first rectangular frame surrounding the target vehicle in the target real-time image; then, based on the position information of the first rectangular frame, intercepting and obtaining a target image in the target real-time image, and carrying out second target detection on the target image to obtain the position information of a second rectangle surrounding a target license plate in the target image; finally, the steering trend of the target vehicle is predicted at least according to the first rectangular frame and the second rectangular frame. In the prediction method, only a first rectangular frame surrounding a target vehicle in a target real-time image is required to be obtained, and a second rectangular frame surrounding a target license plate in the target image is required to be obtained, and then the steering trend of the target vehicle is predicted at least according to the first rectangular frame and the second rectangular frame, so that the target license plate based on the target vehicle is realized, and the steering trend of the target vehicle is predicted. The prediction method of the invention realizes the prediction of the steering trend of the target vehicle through simpler calculation logic, thereby solving the problem of more complex prediction of the steering trend of the automatic driving vehicle in the prior art. The steering trend of the target vehicle is predicted through simpler calculation logic, so that the complexity of calculation of the prediction method is low, the required calculation force is low, and the required cost is low.
In a specific embodiment of the present application, as shown in fig. 2, the target vehicle 101 is in the same lane as the own vehicle 100, and the target vehicle 101 is a vehicle that runs in front of the own vehicle 100.
It should be noted that the steps illustrated in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and that although a logical order is illustrated in the flowcharts, in some cases the steps illustrated or described may be performed in an order other than that illustrated herein.
In order to further more simply predict a steering tendency of the target vehicle, in one embodiment of the present application, predicting the steering tendency of the target vehicle according to at least the first rectangular frame and the second rectangular frame includes: determining a target area in the first rectangular frame by adopting a big data analysis method; predicting that the target vehicle does not have a steering tendency in the case where the second rectangular frame is entirely in the target region; predicting that the target vehicle has a steering tendency in a case where the second rectangular frame is not located in the target region at all; when the second rectangular frame portion is located in the target area, a target coordinate system is constructed with a center point of the target area as a coordinate origin, and a steering tendency of the target vehicle is predicted based on at least a target overlapping distance between the target area and the second rectangular frame in the target coordinate system, the target overlapping distance being a overlapping distance between the target area and the second rectangular frame in a horizontal direction in the target coordinate system. In the embodiment, the big data analysis method is adopted to determine the target area in the first rectangular frame, so that the determined target area is accurate and high in reliability, and whether the target vehicle has a steering trend or not is predicted based on the relative position relation between the target area and the second rectangular frame, so that the accuracy of the prediction result of the steering trend of the target vehicle is ensured.
In a specific embodiment of the present application, as shown in fig. 3, by performing a first target detection on the target real-time image 102, a first rectangular frame 103 surrounding the target vehicle in the target real-time image 102 is obtained. Based on the position information of the first rectangular frame 103 in the target real-time image 102, the target real-time image 102 is cut out to obtain a target image (not shown in fig. 3, but the size of the target image is the same as the size of the first rectangular frame 103). The target image is subjected to second target detection, resulting in a second rectangular frame 105 of the target vehicle surrounding the target vehicle in the target image. Finally, the steering trend of the target vehicle is predicted based on at least the first rectangular frame 103 and the second rectangular frame 105, that is, based on the relative positional relationship between the first rectangular frame 103 and the second rectangular frame 105. The specific process is as follows: based on the method of big data analysis, the target region 104 is predicted in the first rectangular frame 103. If the second rectangular box 105 is completely in the target area 104, the target vehicle is predicted to have no steering tendency, i.e., the target vehicle is in a straight running state. If the second rectangular box 105 is not at all in the target area 104 (this situation is not shown in fig. 3), the target vehicle is predicted to have a steering tendency. If the second rectangular frame 105 is partially located in the target area 104 (i.e. as shown in fig. 5), a target coordinate system 106 is constructed with the center point of the target area 104 (i.e. the o-point in fig. 4) as the origin of coordinates (see fig. 4). The steering tendency of the target vehicle is predicted at least from the target overlap distance of the target region 104 and the second rectangular frame 105 in the target coordinate system 106.
Specifically, in the practical application process, the method of obtaining the target area through big data analysis is not limited, and any feasible method in the prior art may be used to obtain the target area, and the method of obtaining the target area is not limited in this application. The target area is obtained, for example, by classifying the network.
Specifically, in the practical application process, the target coordinate system is not limited to be constructed by taking the center point of the target area as the origin of the coordinate axis, and any corner point of the target area can be adopted to construct the target coordinate system. Of course, the method for calculating the first target ratio and the second target ratio needs to be flexibly adjusted under the condition that the constructed target coordinate systems are different.
In still another embodiment of the present application, as shown in fig. 5, when the second rectangular frame 105 is located in the positive direction of the target coordinate system 106 (i.e., the direction pointed by the arrow in the abscissa of the target coordinate system 106 in fig. 5), predicting the steering tendency of the target vehicle at least according to the target overlapping distance 108 between the target area 104 and the second rectangular frame 105 in the target coordinate system 106 includes: determining a horizontal distance of the target region 104 in a forward direction under the target coordinate system 106 to obtain a first distance 107; calculating the ratio of the target overlapping distance 108 to the first distance 107 to obtain a first target ratio; and if the first target ratio is greater than a predetermined threshold, predicting that the target vehicle has a steering tendency for the first time, and if the first target ratio is less than or equal to the predetermined threshold, predicting that the target vehicle does not have a steering tendency. In the embodiment, calculating the ratio of the target overlapping distance to the first distance to obtain a first target ratio, and under the condition that the first target ratio is larger than a preset threshold value, initially predicting that the target vehicle has a steering trend, and then further predicting the steering trend of the target vehicle based on a plurality of frames of other real-time images, so that the steering trend of the target vehicle is predicted more simply, and the steering trend of the target vehicle is predicted more accurately, namely the reliability of the prediction result of the steering trend of the target vehicle is higher; in the case where the first target ratio is less than or equal to the predetermined threshold value, it is predicted that the target vehicle does not have a steering tendency, further realizing a simpler prediction of whether the target vehicle has a steering tendency.
In another embodiment of the present application, as shown in fig. 6, when the second rectangular frame 105 is located in the negative direction of the target coordinate system 106 (i.e. the direction opposite to the arrow direction in the abscissa of the target coordinate system 106 in fig. 6), the predicting the steering trend of the target vehicle at least according to the target overlapping distance 108 between the target area 104 and the second rectangular frame 105 in the target coordinate system 106 includes: determining an absolute value of a horizontal distance of the target region 104 in a negative direction under the target coordinate system 106 to obtain a second distance 109; calculating the ratio of the target overlapping distance 108 to the second distance 109 to obtain a second target ratio; and if the second target ratio is greater than a predetermined threshold, predicting that the target vehicle has a steering tendency for the first time, and if the second target ratio is less than or equal to the predetermined threshold, predicting that the target vehicle does not have a steering tendency. In the embodiment, calculating the ratio of the target overlapping distance to the second distance to obtain a second target ratio, and under the condition that the second target ratio is larger than a preset threshold value, initially predicting that the target vehicle has a steering trend, and then further predicting the steering trend of the target vehicle based on a plurality of frames of other real-time images, so that the steering trend of the target vehicle is ensured to be predicted more accurately on the basis of more simply predicting the steering trend of the target vehicle; in the case where the second target ratio is less than or equal to the predetermined threshold value, it is predicted that the target vehicle does not have a steering tendency, further realizing a simpler prediction of whether the target vehicle has a steering tendency.
In the practical application process, the size of the predetermined threshold may be flexibly adjusted according to the practical situation, and the size of the predetermined threshold is not limited in the present application.
Specifically, in order to further ensure that the prediction of the steering tendency of the target vehicle is accurate, in still another embodiment of the present application, after the target vehicle is predicted to have the steering tendency for the first time, the prediction method further includes: receiving a plurality of frames of other real-time images, and calculating the first target ratio corresponding to the plurality of frames of other real-time images under the target coordinate system, wherein the acquisition time of the other real-time images is later than that of the target real-time images; and sorting the corresponding first target ratios in time sequence, and determining that the target vehicle has a tendency to turn in the negative direction of the target coordinate system when the first target ratios gradually decrease.
In a specific embodiment of the present application, as shown in fig. 5, in the case that the second rectangular frame 105 is located in the positive direction of the target coordinate system 106, that is, the second rectangular frame 105 is located on the right side of the target coordinate system (determined when a person faces the computer screen), if the first target ratio is greater than the predetermined threshold, the target vehicle is primarily predicted to have a tendency to turn left. In order to accurately predict the target vehicle, a plurality of first target ratios corresponding to other real-time images of the plurality of frames are calculated again with the target coordinate system 106 (i.e., the coordinate system established in the target real-time image) as a reference. And finally, sequencing the corresponding first target ratios according to the sequence of the acquisition moments of the target real-time image and other real-time images of the multiple frames. If the plurality of first target ratios gradually decrease, it is determined that the target vehicle has a tendency to turn left.
In one embodiment of the present application, after the target vehicle is predicted to have a steering tendency for the first time, the prediction method further includes: receiving a plurality of frames of other real-time images, and calculating the second target ratio corresponding to the plurality of frames of other real-time images under the target coordinate system, wherein the acquisition time of the other real-time images is later than that of the target real-time images; and sorting the corresponding second target ratios according to a time sequence, and determining that the target vehicle has a tendency to turn towards the positive direction of the target coordinate system when the second target ratios gradually decrease. In this embodiment, under the condition that the first predicted vehicle has a steering trend, the steering trend of the target vehicle is determined again by adopting the second target ratio corresponding to the other real-time images of the multiple frames, so that the steering trend prediction of the target vehicle is further ensured to be more accurate.
In a specific embodiment of the present application, as shown in fig. 6, in the case that the second rectangular frame 105 is located in the negative direction of the target coordinate system 106, that is, the second rectangular frame 105 is located at the left side of the target coordinate system (determined when a person faces the computer screen), if the second target ratio is greater than the predetermined threshold, the target vehicle is primarily predicted to have a tendency to turn right. In order to accurately predict the target vehicle, a plurality of second target ratios corresponding to other real-time images of the plurality of frames are calculated again with reference to the target coordinate system 106 (i.e., the coordinate system established in the target real-time image). And finally, sequencing the corresponding plurality of second target ratios according to the sequence of the acquisition moments of the target real-time image and other real-time images of the multiple frames. If the plurality of second target ratios gradually decrease, it is determined that the target vehicle has a tendency to turn right.
In another embodiment of the present application, performing a first target detection on a target real-time image sent by a vehicle to obtain location information of a first rectangular frame of the target vehicle in the target real-time image, where the first target detection includes: and performing the first target detection on the target real-time image sent by the vehicle by adopting a first model to obtain the position information of the first rectangular frame of the target vehicle in the target real-time image, wherein the first model is constructed based on a deep learning model. In this embodiment, the first model is used to perform first target detection on the target real-time image, so as to obtain position information of a first rectangular frame surrounding the target vehicle in the target real-time image, that is, the first model is only used to detect the target vehicle in the target real-time image. Because the detection category of the first model is single, the first model is ensured to be lighter, the first model can detect the target real-time image more quickly, and the obtained position information of the first rectangular frame is ensured to be more accurate.
In still another embodiment of the present application, performing a second target detection on the target image to obtain a second rectangular frame of the target license plate in the target image, including: and carrying out the second target detection on the target image by adopting a second model to obtain the second rectangular frame of the target license plate in the target image, wherein the second model is constructed based on a deep learning model. In this embodiment, the second model is used to perform the second target detection on the target image, so as to obtain a second rectangular frame surrounding the target vehicle in the target image, that is, the second model is only used to detect the target license plate in the target graph. Because the detection category of the second model is single, the second model is light and the memory occupied by the second model is small. Meanwhile, the second model can also detect the target image more quickly, the obtained second rectangular frame is more accurate, and further the follow-up prediction of the steering trend of the target vehicle based on the first rectangular frame and the second rectangular frame is more accurate.
The embodiment of the application also provides a device for predicting the vehicle steering, and the device for predicting the vehicle steering can be used for executing the method for predicting the vehicle steering. The following describes a vehicle steering prediction apparatus provided in an embodiment of the present application.
Fig. 7 is a schematic structural view of a vehicle steering prediction apparatus according to an embodiment of the present application. The above prediction apparatus is applied to a server, and as shown in fig. 7, the prediction apparatus includes:
a first detecting unit 10, configured to perform first target detection on a target real-time image sent by a host vehicle, to obtain position information of a first rectangular frame of the target vehicle in the target real-time image, where the target real-time image is an image in front of the host vehicle, the target vehicle is in the same lane as the host vehicle, and the target vehicle is in front of the host vehicle, and the first rectangular frame is a smallest rectangle surrounding the target vehicle;
the second detecting unit 20 is configured to intercept the target real-time image according to the position information of the first rectangular frame, obtain a target image, and perform second target detection on the target image to obtain a second rectangular frame of a target license plate in the target image, where the target license plate is a license plate of the target vehicle, and the second rectangular frame is a minimum rectangle surrounding the target license plate;
And a prediction unit 30 configured to predict a steering tendency of the target vehicle based on at least the first rectangular frame and the second rectangular frame.
In the vehicle steering prediction method, the first detection unit is used for receiving a target real-time image sent by the vehicle and carrying out first target detection on the received target real-time image to obtain the position information of a first rectangular frame surrounding the target vehicle in the target real-time image; the second detection unit is used for obtaining a target image by intercepting in the target real-time image based on the position information of the first rectangular frame, and carrying out second target detection on the target image to obtain the position information of a second rectangle surrounding a target license plate in the target image; the prediction unit is used for predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame. In the prediction device, only the first rectangular frame surrounding the target vehicle in the target real-time image is required to be obtained, and the second rectangular frame surrounding the target license plate in the target image is required to be obtained, and then the steering trend of the target vehicle is predicted at least according to the first rectangular frame and the second rectangular frame, so that the target license plate based on the target vehicle is realized, and the steering trend of the target vehicle is predicted. The prediction device of the application predicts the steering trend of the target vehicle through simpler calculation logic, so that the problem that the steering trend of the automatic driving vehicle is predicted more complicated in the prior art is solved. The steering trend of the target vehicle is predicted through simpler calculation logic, so that the complexity of calculation of the prediction method is low, the required calculation force is low, and the required cost is low.
In a specific embodiment of the present application, as shown in fig. 2, the target vehicle 101 is in the same lane as the own vehicle 100, and the target vehicle 101 is a vehicle that runs in front of the own vehicle 100.
In order to further and more simply predict the steering trend of the target vehicle, in one embodiment of the present application, the prediction unit includes a first determining module, a first prediction module, a second prediction module, and a third prediction module, where the first determining module is configured to determine the target area in the first rectangular frame by using a big data analysis method; the first prediction module is configured to predict that the target vehicle does not have a steering tendency when the second rectangular frame is completely in the target area; the second prediction module is configured to predict that the target vehicle has a steering tendency when the second rectangular frame is not located in the target area at all; the third prediction module is configured to construct a target coordinate system with a center point of the target area as a coordinate origin when the second rectangular frame is not located in the target area, and predict a steering tendency of the target vehicle based on at least a target overlapping distance between the target area and the second rectangular frame in the target coordinate system, where the target overlapping distance is a overlapping distance between the target area and the second rectangular frame in a horizontal direction in the target coordinate system. In the embodiment, the big data analysis method is adopted to determine the target area in the first rectangular frame, so that the determined target area is accurate and high in reliability, and whether the target vehicle has a steering trend or not is predicted based on the relative position relation between the target area and the second rectangular frame, so that the accuracy of the prediction result of the steering trend of the target vehicle is ensured.
In a specific embodiment of the present application, as shown in fig. 3, by performing a first target detection on the target real-time image 102, a first rectangular frame 103 surrounding the target vehicle in the target real-time image 102 is obtained. Based on the position information of the first rectangular frame 103 in the target real-time image 102, the target real-time image 102 is cut out to obtain a target image (not shown in fig. 3, but the size of the target image is the same as the size of the first rectangular frame 103). The target image is subjected to second target detection, resulting in a second rectangular frame 105 of the target vehicle surrounding the target vehicle in the target image. Finally, the steering trend of the target vehicle is predicted based on at least the first rectangular frame 103 and the second rectangular frame 105, that is, based on the relative positional relationship between the first rectangular frame 103 and the second rectangular frame 105. The specific process is as follows: based on the method of big data analysis, the target region 104 is predicted in the first rectangular frame 103. If the second rectangular box 105 is completely in the target area 104, the target vehicle is predicted to have no steering tendency, i.e., the target vehicle is in a straight running state. If the second rectangular box 105 is not at all in the target area 104 (this situation is not shown in fig. 3), the target vehicle is predicted to have a steering tendency. If the second rectangular frame 105 is partially located in the target area 104 (i.e. as shown in fig. 5), a target coordinate system 106 is constructed with the center point of the target area 104 (i.e. the o-point in fig. 4) as the origin of coordinates (see fig. 4). The steering tendency of the target vehicle is predicted at least from the target overlap distance of the target region 104 and the second rectangular frame 105 in the target coordinate system 106.
Specifically, in the practical application process, the method of obtaining the target area through big data analysis is not limited, and any feasible method in the prior art may be used to obtain the target area, and the method of obtaining the target area is not limited in this application. The target area is obtained, for example, by classifying the network.
Specifically, in the practical application process, the target coordinate system is not limited to be constructed by taking the center point of the target area as the origin of the coordinate axis, and any corner point of the target area can be adopted to construct the target coordinate system. Of course, the method for calculating the first target ratio and the second target ratio needs to be flexibly adjusted under the condition that the constructed target coordinate systems are different.
In yet another embodiment of the present application, as shown in fig. 5, in a case where the second rectangular frame 105 is located in the positive direction of the target coordinate system 106 (i.e., the direction pointed by the arrow in the abscissa of the target coordinate system 106 in fig. 5), the third prediction module includes a first determination submodule, a first calculation submodule, and a first prediction submodule, where the first determination submodule is configured to determine the horizontal distance of the target area 104 in the positive direction under the target coordinate system 106, so as to obtain a first distance 107; the first calculating submodule is used for calculating the ratio of the target overlapping distance 108 to the first distance 107 to obtain a first target ratio; the first prediction submodule is used for predicting that the target vehicle has a steering trend for the first time when the first target ratio is larger than a preset threshold value, and predicting that the target vehicle does not have the steering trend when the first target ratio is smaller than or equal to the preset threshold value. In the embodiment, calculating the ratio of the target overlapping distance to the first distance to obtain a first target ratio, and under the condition that the first target ratio is larger than a preset threshold value, initially predicting that the target vehicle has a steering trend, and then further predicting the steering trend of the target vehicle based on a plurality of frames of other real-time images, so that the steering trend of the target vehicle is predicted more simply, and the steering trend of the target vehicle is predicted more accurately, namely the reliability of the prediction result of the steering trend of the target vehicle is higher; in the case where the first target ratio is less than or equal to the predetermined threshold value, it is predicted that the target vehicle does not have a steering tendency, further realizing a simpler prediction of whether the target vehicle has a steering tendency.
In another embodiment of the present application, as shown in fig. 6, in a case where the second rectangular frame 105 is located in the negative direction of the target coordinate system 106 (i.e., in the direction opposite to the direction of the arrow in the abscissa of the target coordinate system 106 in fig. 6), the third prediction module further includes a second determination submodule, a second calculation submodule, and a second prediction submodule, where the second determination submodule is configured to determine an absolute value of the horizontal distance of the target region 104 in the negative direction under the target coordinate system 106, to obtain a second distance 109; the second calculating submodule is used for calculating the ratio of the target overlapping distance 108 to the second distance 109 to obtain a second target ratio; the second prediction submodule is used for predicting that the target vehicle has a steering trend for the first time when the second target ratio is larger than a preset threshold value, and predicting that the target vehicle does not have the steering trend when the second target ratio is smaller than or equal to the preset threshold value. In the embodiment, calculating the ratio of the target overlapping distance to the second distance to obtain a second target ratio, and under the condition that the second target ratio is larger than a preset threshold value, initially predicting that the target vehicle has a steering trend, and then further predicting the steering trend of the target vehicle based on a plurality of frames of other real-time images, so that the steering trend of the target vehicle is ensured to be predicted more accurately on the basis of more simply predicting the steering trend of the target vehicle; in the case where the second target ratio is less than or equal to the predetermined threshold value, it is predicted that the target vehicle does not have a steering tendency, further realizing a simpler prediction of whether the target vehicle has a steering tendency.
In the practical application process, the size of the predetermined threshold may be flexibly adjusted according to the practical situation, and the size of the predetermined threshold is not limited in the present application.
Specifically, in order to further ensure that the prediction of the steering trend of the target vehicle is more accurate, in still another embodiment of the present application, the prediction apparatus further includes a first receiving unit and a first determining unit, where the first receiving unit is configured to receive a plurality of frames of other real-time images after the target vehicle is predicted to have the steering trend for the first time, and calculate the first target ratio corresponding to the plurality of frames of other real-time images in the target coordinate system, where the acquiring time of the other real-time images is later than the target real-time image; the first determining unit is configured to sort the corresponding plurality of first target ratios in time sequence, and determine that the target vehicle has a tendency to turn in a negative direction of the target coordinate system when the plurality of first target ratios gradually decrease.
In a specific embodiment of the present application, as shown in fig. 5, in the case that the second rectangular frame 105 is located in the positive direction of the target coordinate system 106, that is, the second rectangular frame 105 is located on the right side of the target coordinate system (determined when a person faces the computer screen), if the first target ratio is greater than the predetermined threshold, the target vehicle is primarily predicted to have a tendency to turn left. In order to accurately predict the target vehicle, a plurality of first target ratios corresponding to other real-time images of the plurality of frames are calculated again with the target coordinate system 106 (i.e., the coordinate system established in the target real-time image) as a reference. And finally, sequencing the corresponding first target ratios according to the sequence of the acquisition moments of the target real-time image and other real-time images of the multiple frames. If the plurality of first target ratios gradually decrease, it is determined that the target vehicle has a tendency to turn left.
In an embodiment of the present application, the predicting device further includes a second receiving unit and a second determining unit, where the second receiving unit is configured to receive a plurality of frames of other real-time images after predicting that the target vehicle has a steering tendency for the first time, and calculate the second target ratio corresponding to the plurality of frames of other real-time images in the target coordinate system, where an acquisition time of the other real-time images is later than that of the target real-time image; the second determining unit is configured to sort a plurality of second target ratios corresponding to each other in time sequence, and determine that the target vehicle has a tendency to turn in a forward direction of the target coordinate system when the plurality of second target ratios gradually decrease. In this embodiment, under the condition that the first predicted vehicle has a steering trend, the steering trend of the target vehicle is determined again by adopting the second target ratio corresponding to the other real-time images of the multiple frames, so that the steering trend prediction of the target vehicle is further ensured to be more accurate.
In a specific embodiment of the present application, as shown in fig. 6, in the case that the second rectangular frame 105 is located in the negative direction of the target coordinate system 106, that is, the second rectangular frame 105 is located at the left side of the target coordinate system (determined when a person faces the computer screen), if the second target ratio is greater than the predetermined threshold, the target vehicle is primarily predicted to have a tendency to turn right. In order to accurately predict the target vehicle, a plurality of second target ratios corresponding to other real-time images of the plurality of frames are calculated again with reference to the target coordinate system 106 (i.e., the coordinate system established in the target real-time image). And finally, sequencing the corresponding plurality of second target ratios according to the sequence of the acquisition moments of the target real-time image and other real-time images of the multiple frames. If the plurality of second target ratios gradually decrease, it is determined that the target vehicle has a tendency to turn right.
In another embodiment of the present application, the first detection unit includes a first detection module configured to perform the first target detection on the target real-time image sent by the host vehicle by using a first model to obtain position information of the first rectangular frame of the target vehicle in the target real-time image, where the first model is constructed based on a deep learning model. In this embodiment, the first model is used to perform first target detection on the target real-time image, so as to obtain position information of a first rectangular frame surrounding the target vehicle in the target real-time image, that is, the first model is only used to detect the target vehicle in the target real-time image. Because the detection category of the first model is single, the first model is ensured to be lighter, the first model can detect the target real-time image more quickly, and the obtained position information of the first rectangular frame is ensured to be more accurate.
In still another embodiment of the present application, the second detection unit includes a second detection module, configured to perform the second target detection on the target image by using a second model to obtain the second rectangular frame of the target license plate in the target image, where the second model is constructed based on a deep learning model. In this embodiment, the second model is used to perform the second target detection on the target image, so as to obtain a second rectangular frame surrounding the target vehicle in the target image, that is, the second model is only used to detect the target license plate in the target graph. Because the detection category of the second model is single, the second model is light and the memory occupied by the second model is small. Meanwhile, the second model can also detect the target image more quickly, the obtained second rectangular frame is more accurate, and further the follow-up prediction of the steering trend of the target vehicle based on the first rectangular frame and the second rectangular frame is more accurate.
The vehicle steering prediction device comprises a processor and a memory, wherein the first detection unit, the second detection unit, the prediction unit and the like are stored in the memory as program units, and the processor executes the program units stored in the memory to realize corresponding functions.
The processor includes a kernel, and the kernel fetches the corresponding program unit from the memory. The kernel can be provided with one or more than one kernel, and the problem that the steering trend of the automatic driving vehicle is predicted to be complex in the prior art is solved by adjusting the kernel parameters.
The memory may include volatile memory, random Access Memory (RAM), and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM), among other forms in computer readable media, the memory including at least one memory chip.
An embodiment of the present invention provides a computer-readable storage medium having stored thereon a program that, when executed by a processor, implements the above-described vehicle steering prediction method.
The embodiment of the invention provides a processor which is used for running a program, wherein the program runs to execute the method for predicting the steering of the vehicle.
The embodiment of the invention provides equipment, which comprises a processor, a memory and a program stored in the memory and capable of running on the processor, wherein the processor realizes at least the following steps when executing the program:
step S101, performing first target detection on a target real-time image sent by a vehicle to obtain position information of a first rectangular frame of the target vehicle in the target real-time image, wherein the target real-time image is an image in front of the vehicle, the target vehicle and the vehicle are in the same lane, the target vehicle is in front of the vehicle, and the first rectangular frame is a smallest rectangle surrounding the target vehicle;
step S102, intercepting the target real-time image according to the position information of the first rectangular frame to obtain a target image, and carrying out second target detection on the target image to obtain a second rectangular frame of a target license plate in the target image, wherein the target license plate is the license plate of the target vehicle, and the second rectangular frame is the smallest rectangle surrounding the target license plate;
and step S103, predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame.
The device herein may be a server, PC, PAD, cell phone, etc.
The present application also provides a computer program product adapted to perform a program initialized with at least the following method steps when executed on a data processing device:
step S101, performing first target detection on a target real-time image sent by a vehicle to obtain position information of a first rectangular frame of the target vehicle in the target real-time image, wherein the target real-time image is an image in front of the vehicle, the target vehicle and the vehicle are in the same lane, the target vehicle is in front of the vehicle, and the first rectangular frame is a smallest rectangle surrounding the target vehicle;
step S102, intercepting the target real-time image according to the position information of the first rectangular frame to obtain a target image, and carrying out second target detection on the target image to obtain a second rectangular frame of a target license plate in the target image, wherein the target license plate is the license plate of the target vehicle, and the second rectangular frame is the smallest rectangle surrounding the target license plate;
and step S103, predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame.
In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units may be a logic function division, and there may be another division manner when actually implemented, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described above as separate components may or may not be physically separate, and components shown as units may or may not be physical units, may be located in one place, or may be distributed over a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units described above, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied in essence or a part contributing to the prior art or all or part of the technical solution in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the above-mentioned method of the various embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:
1) In the vehicle steering prediction method, firstly, a target real-time image sent by a vehicle of a host vehicle is received, and the received target real-time image is subjected to first target detection to obtain the position information of a first rectangular frame surrounding the target vehicle in the target real-time image; then, based on the position information of the first rectangular frame, intercepting and obtaining a target image in the target real-time image, and carrying out second target detection on the target image to obtain the position information of a second rectangle surrounding a target license plate in the target image; finally, the steering trend of the target vehicle is predicted at least according to the first rectangular frame and the second rectangular frame. In the prediction method, only a first rectangular frame surrounding a target vehicle in a target real-time image is required to be obtained, and a second rectangular frame surrounding a target license plate in the target image is required to be obtained, and then the steering trend of the target vehicle is predicted at least according to the first rectangular frame and the second rectangular frame, so that the target license plate based on the target vehicle is realized, and the steering trend of the target vehicle is predicted. The prediction method of the invention realizes the prediction of the steering trend of the target vehicle through simpler calculation logic, thereby solving the problem of more complex prediction of the steering trend of the automatic driving vehicle in the prior art. The steering trend of the target vehicle is predicted through simpler calculation logic, so that the complexity of calculation of the prediction method is low, the required calculation force is low, and the required cost is low.
2) In the vehicle steering prediction method, a first detection unit is used for receiving a target real-time image sent by a vehicle of a host vehicle, and carrying out first target detection on the received target real-time image to obtain position information of a first rectangular frame surrounding the target vehicle in the target real-time image; the second detection unit is used for obtaining a target image by intercepting in the target real-time image based on the position information of the first rectangular frame, and carrying out second target detection on the target image to obtain the position information of a second rectangle surrounding a target license plate in the target image; the prediction unit is used for predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame. In the prediction device, only the first rectangular frame surrounding the target vehicle in the target real-time image is required to be obtained, and the second rectangular frame surrounding the target license plate in the target image is required to be obtained, and then the steering trend of the target vehicle is predicted at least according to the first rectangular frame and the second rectangular frame, so that the target license plate based on the target vehicle is realized, and the steering trend of the target vehicle is predicted. The prediction device of the application predicts the steering trend of the target vehicle through simpler calculation logic, so that the problem that the steering trend of the automatic driving vehicle is predicted more complicated in the prior art is solved. The steering trend of the target vehicle is predicted through simpler calculation logic, so that the complexity of calculation of the prediction method is low, the required calculation force is low, and the required cost is low.
The foregoing description is only of the preferred embodiments of the present application and is not intended to limit the same, but rather, various modifications and variations may be made by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims (10)

1. A prediction method of vehicle steering, the prediction method being applied in a server, the prediction method comprising:
performing first target detection on a target real-time image sent by a vehicle to obtain position information of a first rectangular frame of the target vehicle in the target real-time image, wherein the target real-time image is an image in front of the vehicle, the target vehicle and the vehicle are in the same lane, the target vehicle is in front of the vehicle, and the first rectangular frame is a smallest rectangle surrounding the target vehicle;
intercepting the target real-time image according to the position information of the first rectangular frame to obtain a target image, and performing second target detection on the target image to obtain a second rectangular frame of a target license plate in the target image, wherein the target license plate is a license plate of the target vehicle, and the second rectangular frame is a minimum rectangle surrounding the target license plate;
And predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame.
2. The prediction method according to claim 1, characterized in that predicting the steering tendency of the target vehicle based on at least the first rectangular frame and the second rectangular frame includes:
determining a target area in the first rectangular frame by adopting a big data analysis method;
predicting that the target vehicle does not have a steering tendency in the case where the second rectangular frame is entirely in the target region;
predicting that the target vehicle has a steering tendency in a case where the second rectangular frame is not at all in the target region;
and under the condition that the second rectangular frame part is positioned in the target area, constructing a target coordinate system by taking the central point of the target area as the coordinate origin, and predicting the steering trend of the target vehicle at least according to the target overlapping distance of the target area and the second rectangular frame under the target coordinate system, wherein the target overlapping distance is the overlapping distance of the target area and the second rectangular frame in the horizontal direction under the target coordinate system.
3. The prediction method according to claim 2, wherein predicting the steering tendency of the target vehicle based on at least a target overlap distance of the target region and the second rectangular frame in the target coordinate system in a case where the second rectangular frame is located in a positive direction of the target coordinate system, comprises:
determining the horizontal distance of the target area in the positive direction under the target coordinate system to obtain a first distance;
calculating the ratio of the target overlapping distance to the first distance to obtain a first target ratio;
and if the first target ratio is smaller than or equal to the preset threshold value, predicting that the target vehicle does not have steering trend.
4. The prediction method according to claim 2, wherein predicting the steering tendency of the target vehicle based on at least the target overlap distance of the target region and the second rectangular frame in the target coordinate system in the case where the second rectangular frame is located in the negative direction of the target coordinate system, comprises:
Determining the absolute value of the horizontal distance of the target area in the negative direction under the target coordinate system to obtain a second distance;
calculating the ratio of the target overlapping distance to the second distance to obtain a second target ratio;
and if the second target ratio is smaller than or equal to the preset threshold value, predicting that the target vehicle does not have steering trend.
5. The prediction method according to claim 3, characterized in that, after the target vehicle is predicted to have a steering tendency for the first time, the prediction method further comprises:
receiving a plurality of frames of other real-time images, and calculating the first target ratio corresponding to the frames of other real-time images under the target coordinate system, wherein the acquisition time of the other real-time images is later than that of the target real-time images;
and sequencing the corresponding first target ratios in time sequence, and determining that the target vehicle has a tendency to turn towards the negative direction of the target coordinate system under the condition that the first target ratios gradually decrease.
6. The prediction method according to claim 4, characterized in that, after the target vehicle is predicted to have a steering tendency for the first time, the prediction method further comprises:
receiving a plurality of frames of other real-time images, and calculating the second target ratio corresponding to the frames of other real-time images under the target coordinate system, wherein the acquisition time of the other real-time images is later than that of the target real-time images;
and sequencing the corresponding second target ratios according to a time sequence, and determining that the target vehicle has a tendency to turn towards the positive direction of the target coordinate system under the condition that the second target ratios gradually decrease.
7. The prediction method according to any one of claims 1 to 6, wherein performing first target detection on a target real-time image transmitted from a host vehicle to obtain position information of a first rectangular frame of the target vehicle in the target real-time image includes:
and carrying out first target detection on the target real-time image sent by the vehicle by adopting a first model to acquire the position information of the first rectangular frame of the target vehicle in the target real-time image, wherein the first model is constructed based on a deep learning model.
8. The prediction method according to any one of claims 1 to 6, wherein performing a second target detection on the target image to obtain a second rectangular frame of a target license plate in the target image includes:
and carrying out second target detection on the target image by adopting a second model to obtain the second rectangular frame of the target license plate in the target image, wherein the second model is constructed based on a deep learning model.
9. A prediction apparatus for vehicle steering, the prediction apparatus being applied in a server, the prediction apparatus comprising:
the first detection unit is used for carrying out first target detection on a target real-time image sent by a vehicle to obtain position information of a first rectangular frame of the target vehicle in the target real-time image, wherein the target real-time image is an image in front of the vehicle, the target vehicle and the vehicle are in the same lane, the target vehicle is in front of the vehicle, and the first rectangular frame is a smallest rectangle surrounding the target vehicle;
the second detection unit is used for intercepting the target real-time image according to the position information of the first rectangular frame to obtain a target image, and carrying out second target detection on the target image to obtain a second rectangular frame of a target license plate in the target image, wherein the target license plate is a license plate of the target vehicle, and the second rectangular frame is a minimum rectangle surrounding the target license plate;
And the prediction unit is used for predicting the steering trend of the target vehicle at least according to the first rectangular frame and the second rectangular frame.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium includes a stored program, wherein the program executes the vehicle steering prediction method according to any one of claims 1 to 8.
CN202211641437.4A 2022-12-20 2022-12-20 Method and device for predicting vehicle steering and computer readable storage medium Pending CN116311144A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211641437.4A CN116311144A (en) 2022-12-20 2022-12-20 Method and device for predicting vehicle steering and computer readable storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211641437.4A CN116311144A (en) 2022-12-20 2022-12-20 Method and device for predicting vehicle steering and computer readable storage medium

Publications (1)

Publication Number Publication Date
CN116311144A true CN116311144A (en) 2023-06-23

Family

ID=86787575

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211641437.4A Pending CN116311144A (en) 2022-12-20 2022-12-20 Method and device for predicting vehicle steering and computer readable storage medium

Country Status (1)

Country Link
CN (1) CN116311144A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116778466A (en) * 2023-07-20 2023-09-19 哪吒港航智慧科技(上海)有限公司 Large-angle license plate recognition method, device, equipment and storage medium

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116778466A (en) * 2023-07-20 2023-09-19 哪吒港航智慧科技(上海)有限公司 Large-angle license plate recognition method, device, equipment and storage medium

Similar Documents

Publication Publication Date Title
US20210213961A1 (en) Driving scene understanding
CN112330601A (en) Parking detection method, device, equipment and medium based on fisheye camera
US11126875B2 (en) Method and device of multi-focal sensing of an obstacle and non-volatile computer-readable storage medium
CN114170826B (en) Automatic driving control method and device, electronic device and storage medium
CN116311144A (en) Method and device for predicting vehicle steering and computer readable storage medium
CN111753702A (en) Target detection method, device and equipment
CN112634201A (en) Target detection method and device and electronic equipment
CN113428177B (en) Vehicle control method, device, equipment and storage medium
CN116863124B (en) Vehicle attitude determination method, controller and storage medium
CN115792945B (en) Floating obstacle detection method and device, electronic equipment and storage medium
CN109214256A (en) A kind of communication chart object detection method, device and vehicle
CN109188419B (en) Method and device for detecting speed of obstacle, computer equipment and storage medium
US20220122341A1 (en) Target detection method and apparatus, electronic device, and computer storage medium
CN110555344B (en) Lane line recognition method, lane line recognition device, electronic device, and storage medium
CN115830562A (en) Method for determining lane information, computer device, and medium
US11989957B2 (en) Apparatus for controlling object tracking and method therefor
CN111177878A (en) Method, device and terminal for screening derivative simulation scenes
CN109740518B (en) Method and device for determining object in video
CN111199179B (en) Target object tracking method, terminal equipment and medium
CN111369373A (en) Vehicle interior damage determination method and device
CN111539361B (en) Noise identification method, device, storage medium, processor and carrier
CN112633214B (en) Vehicle identification method and device
CN113158864B (en) Method and device for determining included angle between truck head and trailer
CN115303280B (en) Redundant low-delay vehicle driving road identification method and device
US20220012368A1 (en) Biometric feature attack objects and biometric feature attack test systems

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination