CN112983085A - Parking space line identification method based on vision - Google Patents
Parking space line identification method based on vision Download PDFInfo
- Publication number
- CN112983085A CN112983085A CN202110480377.1A CN202110480377A CN112983085A CN 112983085 A CN112983085 A CN 112983085A CN 202110480377 A CN202110480377 A CN 202110480377A CN 112983085 A CN112983085 A CN 112983085A
- Authority
- CN
- China
- Prior art keywords
- parking space
- image
- coordinates
- slot position
- parking
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 21
- 240000004282 Grewia occidentalis Species 0.000 claims description 14
- 238000001514 detection method Methods 0.000 claims description 7
- 230000009466 transformation Effects 0.000 claims description 6
- 230000004888 barrier function Effects 0.000 claims description 3
- 230000001788 irregular Effects 0.000 claims description 3
- 238000005286 illumination Methods 0.000 abstract description 2
- 230000000007 visual effect Effects 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 3
- 238000000605 extraction Methods 0.000 description 2
- 206010063659 Aversion Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
Images
Classifications
-
- E—FIXED CONSTRUCTIONS
- E04—BUILDING
- E04H—BUILDINGS OR LIKE STRUCTURES FOR PARTICULAR PURPOSES; SWIMMING OR SPLASH BATHS OR POOLS; MASTS; FENCING; TENTS OR CANOPIES, IN GENERAL
- E04H6/00—Buildings for parking cars, rolling-stock, aircraft, vessels or like vehicles, e.g. garages
- E04H6/42—Devices or arrangements peculiar to garages, not covered elsewhere, e.g. securing devices, safety devices, monitoring and operating schemes; centering devices
- E04H6/426—Parking guides
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/144—Image acquisition using a slot moved over the image; using discrete sensing elements at predetermined points; using automatic curve following means
Landscapes
- Engineering & Computer Science (AREA)
- Architecture (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a visual-based parking space line identification method, which relates to the field of parking space identification of automatic parking, and comprises the following steps: acquiring a look-around image, extracting the characteristics of the look-around image to obtain parking space characteristics, identifying a slot position point and a parking space type, and calculating the center coordinate of a slot position point positioning frame; if a plurality of parking spaces exist in the panoramic image, pairwise matching is carried out on the slot position points according to the fact that the center coordinates of two slot position point positioning frames of the same parking space are located in the parking space positioning frames; calculating the coordinates of the two remaining angular points of the same parking space according to the type of the parking space and the center coordinates of the two matched slot position point positioning frames; processing the image according to the four-angle-point coordinates of the parking space to obtain a parking space image, extracting image characteristics, and judging whether an obstacle exists in the parking space; if no obstacle exists in the parking space, the parking space can stop, and coordinates of four corner points of the parking space can be output. The invention can accurately detect the parking space when facing the complex parking space and the complex illumination environment.
Description
Technical Field
The invention relates to the field of parking space identification of automatic parking, in particular to a parking space line identification method based on vision.
Background
With the rapid development of automobile technology and the improvement of automobile popularity, the space of roadside parking spaces is smaller and smaller. It is difficult for many beginners to perform parking operations in a narrow space, and even a skilled driver has a feeling of aversion to the parking operations, which is a problem to be solved by the automatic parking technique.
The automatic parking system comprises three parts, namely target parking space designation, path planning and path tracking. The target parking space designation method can be divided into four categories, namely, based on a human-computer interaction interface, based on a parking space marking line, based on vehicles on two sides of a parking space and based on infrastructure. The sensors used for detecting the parking spaces mainly comprise a visual sensor and a distance measuring sensor.
The existing products detect the effective parking space by adopting an ultrasonic sensor, but the most attractive method for the driver is a method based on machine vision or deep learning, because the parking environment and the parking process can be visualized, the driver has more safety. Meanwhile, the automatic parking system is mainly applied to urban areas, and the parking space detection method based on the parking space marking lines is most economical and effective.
Disclosure of Invention
The purpose of the invention is as follows: aiming at the problem of inaccurate parking space detection in the existing automatic parking system, the invention provides a method for detecting a vehicle position line in the field of automatic parking, which can accurately detect parking spaces in the face of complex parking spaces and complex illumination environments.
The technical scheme is as follows: in order to realize the purpose of the invention, the technical scheme adopted by the invention is as follows: a parking space line identification method based on vision comprises the following steps:
the method comprises the following steps: acquiring a look-around image, extracting the characteristics of the look-around image to obtain parking space characteristics, identifying a slot position point and a parking space type, and calculating the center coordinate of a slot position point positioning frame;
step two: if a plurality of parking spaces exist in the panoramic image, pairwise matching is carried out on the slot position points according to the fact that the center coordinates of two slot position point positioning frames of the same parking space are located in the parking space positioning frames;
step three: calculating the coordinates of the two remaining angular points of the same parking space according to the type of the parking space and the center coordinates of the two matched slot position point positioning frames;
step four: processing the image according to the four-corner coordinates of the parking space to obtain a parking space image, and extracting the characteristics of the parking space image;
step five: and judging whether an obstacle exists in the parking space or not according to the extracted image characteristics, if no obstacle exists in the parking space, stopping the parking space, and outputting coordinates of four corner points of the parking space.
And further, according to the ratio of the image resolution and the actual physical distance, the ratio of the distance between the slot position points of the corresponding parking place type and the length and width of the parking place, and the center coordinates of the two matched slot position point positioning frames, and by combining the parallelogram principle, the coordinates of the two remaining angle points of the same parking place are calculated in an inference manner.
Further, in the fourth step, the image is cut according to the coordinates of the four corner points of the parking space to obtain an irregular quadrilateral area image, and the image is subjected to perspective transformation to obtain a parking space image.
The invention also provides a parking space line identification system, which comprises:
the parking space identification module is used for acquiring the look-around image, extracting the image characteristics, identifying the slot position and the parking space type and calculating the center coordinate of the slot position frame;
the slot position point data processing module is used for pairing the slot position points in pairs according to the central coordinates of two slot position point positioning frames of the same parking space in the parking space positioning frames; simultaneously calculating the coordinates of the two remaining angular points of the same parking space;
the parking space stopping detection module is used for cutting and perspective transformation of the image according to the parking space four-corner point coordinates, extracting image characteristics, judging whether barriers exist in the parking space or not and outputting the parking space four-corner point coordinates.
Has the advantages that: compared with the prior art, the technical scheme of the invention has the following beneficial technical effects:
the method provided by the invention is based on processing the look-around image and judging whether the parking spaces, the positions of the parking space positioning points and the types of the parking spaces exist in the image.
Drawings
FIG. 1 is a flow chart of the algorithm of the present invention;
FIG. 2 is a schematic view of parking space detection;
FIG. 3 is a schematic of slot point matching.
Detailed Description
The technical solution of the present invention is further described below with reference to the accompanying drawings and examples.
As shown in fig. 1, an embodiment of the present invention provides a parking space line identification method based on vision, including the following steps:
the method comprises the following steps: firstly, acquiring real-time images of a plurality of cameras for splicing to obtain a panoramic image; performing feature extraction on the panoramic image to obtain parking space features, identifying a slot position point and a parking space type, and calculating a central coordinate of a slot position point positioning frame; the parking space types include: horizontal parking spaces, 60-degree inclined parking spaces, 120-degree inclined parking spaces and vertical parking spaces.
Step two: if the surrounding view image is recognized to have a plurality of parking spaces, the surrounding view image indicates that a plurality of slot positions exist, and the slot positions are paired pairwise according to the fact that the center coordinates of two slot position positioning frames of the same parking space are located in the parking space positioning frames.
Step three: calculating the coordinates of the two remaining angular points of the same parking space according to the parking space type (the parking space type corresponds to the parking space angle) and the center coordinates of the two matched slot position positioning frames;
specifically, the coordinates of the two remaining angular points of the same parking space are calculated by inference according to the proportion of the image resolution and the actual physical distance, the proportion of the distance between the slot position points of the corresponding parking space type and the length and width of the parking space, and the center coordinates of the two matched slot position point positioning frames in combination with a parallelogram principle.
Step four: the image is cut according to the coordinates of the four corner points of the parking space to obtain an irregular quadrilateral area image, perspective transformation is carried out on the image to obtain a parking space image with the resolution of 120x64, and feature extraction is carried out on the image.
Step five: and judging whether an obstacle exists in the parking space or not according to the extracted image characteristics, if no obstacle exists in the parking space, stopping the parking space, and outputting coordinates of four corner points of the parking space.
The embodiment of the present invention further provides a parking space line identification system, including:
the parking space identification module is used for acquiring the look-around image, extracting the image characteristics, identifying the slot position and the parking space type and calculating the center coordinate of the slot position frame;
specifically, the parking space identification module processes 608x608 resolution images on the basis of a YOLOV4 open source algorithm to identify slot sites, horizontal parking spaces, 60-degree inclined parking spaces, 120-degree inclined parking spaces and vertical parking spaces;
the slot position point data processing module is used for pairing the slot position points in pairs according to the central coordinates of two slot position point positioning frames of the same parking space in the parking space positioning frames; simultaneously calculating the coordinates of the two remaining angular points of the same parking space;
as shown in fig. 3, the small rectangular frame is a positioning frame of a slot position, the large rectangular frame is a positioning frame of a parking space type, and the center coordinates of the small rectangular frame are inside the large rectangular frame, so that the slot positions of a plurality of parking spaces in the image can be paired two by two;
the parking space stopping detection module is used for cutting and carrying out perspective transformation on the image according to the coordinates of the four corner points of the parking space, extracting the characteristics of the image, judging whether barriers exist in the parking space or not and outputting the coordinates of the four corner points of the parking space capable of stopping;
specifically, the parking space parking detection module processes an image with a resolution of 120x64 on the basis of an AlexNet open source algorithm to obtain image features, classifies the image, and judges whether the parking space can be parked or not.
As shown in fig. 2, the system of the present invention detects four points (angular points encircled by circles) on two parking spaces beside a vehicle, matches the points, the points encircled by the two circles on the upper side form an effective parking space, and the points encircled by the two circles on the lower side form an effective parking space; meanwhile, detecting the type of the parking space, and judging that the parking space belongs to a vertical parking space; and (4) deducing the coordinates (circled by a rectangular frame) of the remaining two angular points according to the parking place type and the coordinates of the positioning points in the circular ring, thus obtaining the complete parking place coordinates.
Claims (4)
1. A parking space line identification method based on vision is characterized by comprising the following steps:
the method comprises the following steps: acquiring a look-around image, extracting the characteristics of the look-around image to obtain parking space characteristics, identifying a slot position point and a parking space type, and calculating the center coordinate of a slot position point positioning frame;
step two: if a plurality of parking spaces exist in the panoramic image, pairwise matching is carried out on the slot position points according to the fact that the center coordinates of two slot position point positioning frames of the same parking space are located in the parking space positioning frames;
step three: calculating the coordinates of the two remaining angular points of the same parking space according to the type of the parking space and the center coordinates of the two matched slot position point positioning frames;
step four: processing the image according to the four-corner coordinates of the parking space to obtain a parking space image, and extracting the characteristics of the parking space image;
step five: and judging whether an obstacle exists in the parking space or not according to the extracted image characteristics, if no obstacle exists in the parking space, stopping the parking space, and outputting coordinates of four corner points of the parking space.
2. The vision-based parking space line identification method according to claim 1, wherein the three steps are based on the ratio of the image resolution to the actual physical distance, the distance between the slot positions of the corresponding parking space type, the ratio of the length and the width of the parking space, and the center coordinates of the matched two slot position positioning frames, and the parallelogram principle is combined to calculate the coordinates of the remaining two angular points of the same parking space in an inference manner.
3. The vision-based parking space line identification method according to claim 1 or 2, wherein in the four steps, the image is cut according to four-corner coordinates of the parking space to obtain an irregular quadrilateral area image, and the image is subjected to perspective transformation to obtain a parking space image.
4. The utility model provides a parking stall line identification system which characterized in that, the system includes:
the parking space identification module is used for acquiring the look-around image, extracting the image characteristics, identifying the slot position and the parking space type and calculating the center coordinate of the slot position frame;
the slot position point data processing module is used for pairing the slot position points in pairs according to the central coordinates of two slot position point positioning frames of the same parking space in the parking space positioning frames; simultaneously calculating the coordinates of the two remaining angular points of the same parking space;
the parking space stopping detection module is used for cutting and perspective transformation of the image according to the parking space four-corner point coordinates, extracting image characteristics, judging whether barriers exist in the parking space or not and outputting the parking space four-corner point coordinates.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110480377.1A CN112983085A (en) | 2021-04-30 | 2021-04-30 | Parking space line identification method based on vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110480377.1A CN112983085A (en) | 2021-04-30 | 2021-04-30 | Parking space line identification method based on vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112983085A true CN112983085A (en) | 2021-06-18 |
Family
ID=76336862
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110480377.1A Pending CN112983085A (en) | 2021-04-30 | 2021-04-30 | Parking space line identification method based on vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112983085A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113627276A (en) * | 2021-07-20 | 2021-11-09 | 的卢技术有限公司 | Method and device for detecting parking space |
CN113792638A (en) * | 2021-09-07 | 2021-12-14 | 上海电力大学 | Thermal power plant rain drainage port pollutant identification method based on parallelgram-Yolov 4 |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107424116A (en) * | 2017-07-03 | 2017-12-01 | 浙江零跑科技有限公司 | Position detecting method of parking based on side ring depending on camera |
CN109435942A (en) * | 2018-10-31 | 2019-03-08 | 合肥工业大学 | A kind of parking stall line parking stall recognition methods and device based on information fusion |
CN110490172A (en) * | 2019-08-27 | 2019-11-22 | 北京茵沃汽车科技有限公司 | Information merges parking stall position compensation method, the system, device, medium parked |
CN110861639A (en) * | 2019-11-28 | 2020-03-06 | 安徽江淮汽车集团股份有限公司 | Parking information fusion method and device, electronic equipment and storage medium |
CN111178236A (en) * | 2019-12-27 | 2020-05-19 | 清华大学苏州汽车研究院(吴江) | Parking space detection method based on deep learning |
CN111191485A (en) * | 2018-11-14 | 2020-05-22 | 广州汽车集团股份有限公司 | Parking space detection method and system and automobile |
CN111845723A (en) * | 2020-08-05 | 2020-10-30 | 北京四维智联科技有限公司 | Full-automatic parking method and system |
-
2021
- 2021-04-30 CN CN202110480377.1A patent/CN112983085A/en active Pending
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107424116A (en) * | 2017-07-03 | 2017-12-01 | 浙江零跑科技有限公司 | Position detecting method of parking based on side ring depending on camera |
CN109435942A (en) * | 2018-10-31 | 2019-03-08 | 合肥工业大学 | A kind of parking stall line parking stall recognition methods and device based on information fusion |
CN111191485A (en) * | 2018-11-14 | 2020-05-22 | 广州汽车集团股份有限公司 | Parking space detection method and system and automobile |
CN110490172A (en) * | 2019-08-27 | 2019-11-22 | 北京茵沃汽车科技有限公司 | Information merges parking stall position compensation method, the system, device, medium parked |
CN110861639A (en) * | 2019-11-28 | 2020-03-06 | 安徽江淮汽车集团股份有限公司 | Parking information fusion method and device, electronic equipment and storage medium |
CN111178236A (en) * | 2019-12-27 | 2020-05-19 | 清华大学苏州汽车研究院(吴江) | Parking space detection method based on deep learning |
CN111845723A (en) * | 2020-08-05 | 2020-10-30 | 北京四维智联科技有限公司 | Full-automatic parking method and system |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113627276A (en) * | 2021-07-20 | 2021-11-09 | 的卢技术有限公司 | Method and device for detecting parking space |
CN113792638A (en) * | 2021-09-07 | 2021-12-14 | 上海电力大学 | Thermal power plant rain drainage port pollutant identification method based on parallelgram-Yolov 4 |
CN113792638B (en) * | 2021-09-07 | 2024-03-05 | 上海电力大学 | Thermal power plant rain exhaust pollutant identification method based on Parallelogram-YOLOv4 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7046822B1 (en) | Method of detecting objects within a wide range of a road vehicle | |
CN107738612B (en) | Automatic parking space detection and identification system based on panoramic vision auxiliary system | |
CN108444390B (en) | Unmanned automobile obstacle identification method and device | |
CN101976460B (en) | Generating method of virtual view image of surveying system of vehicular multi-lens camera | |
US8154594B2 (en) | Mobile peripheral monitor | |
US8331653B2 (en) | Object detector | |
CN103559791B (en) | A kind of vehicle checking method merging radar and ccd video camera signal | |
CN109409202A (en) | Robustness method for detecting lane lines based on dynamic area-of-interest | |
CN107341454A (en) | The detection method and device of barrier, electronic equipment in a kind of scene | |
JP4171501B2 (en) | Vehicle periphery monitoring device | |
CN107392103A (en) | The detection method and device of road surface lane line, electronic equipment | |
CN107650906B (en) | Intelligent parking system and control method thereof | |
CN110909705B (en) | Road side parking space sensing method and system based on vehicle-mounted camera | |
CN112983085A (en) | Parking space line identification method based on vision | |
CN111856417B (en) | Performance analysis method, device, terminal and storage medium of vehicle millimeter wave radar | |
TWI468647B (en) | Obstacle detection system and obstacle detection method thereof | |
WO2007013489A1 (en) | Parking assistance device | |
CN107796373B (en) | Distance measurement method based on monocular vision of front vehicle driven by lane plane geometric model | |
CN108594244B (en) | Obstacle recognition transfer learning method based on stereoscopic vision and laser radar | |
CN112132896A (en) | Trackside equipment state detection method and system | |
JP2002024986A (en) | Pedestrian detector | |
CN108364476B (en) | Method and device for acquiring Internet of vehicles information | |
WO2022033484A1 (en) | Garbage recognition method and system for intelligent sweeper | |
CN109827516B (en) | Method for measuring distance through wheel | |
Zhang et al. | An improved parking space recognition algorithm based on panoramic vision |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210618 |
|
RJ01 | Rejection of invention patent application after publication |