CN113724527A - Parking space management method - Google Patents
Parking space management method Download PDFInfo
- Publication number
- CN113724527A CN113724527A CN202110999594.1A CN202110999594A CN113724527A CN 113724527 A CN113724527 A CN 113724527A CN 202110999594 A CN202110999594 A CN 202110999594A CN 113724527 A CN113724527 A CN 113724527A
- Authority
- CN
- China
- Prior art keywords
- parking space
- image
- area
- parking
- space state
- 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
- 238000007726 management method Methods 0.000 title claims abstract description 32
- 238000001514 detection method Methods 0.000 claims abstract description 26
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000007781 pre-processing Methods 0.000 claims abstract description 15
- 238000013135 deep learning Methods 0.000 claims abstract description 10
- 238000013145 classification model Methods 0.000 claims description 12
- 239000011159 matrix material Substances 0.000 claims description 11
- 230000011218 segmentation Effects 0.000 claims description 9
- 230000009466 transformation Effects 0.000 claims description 8
- 238000005070 sampling Methods 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 10
- 238000012544 monitoring process Methods 0.000 abstract description 6
- 238000013523 data management Methods 0.000 abstract description 5
- 230000003993 interaction Effects 0.000 abstract description 4
- 230000008569 process Effects 0.000 abstract description 4
- 238000012545 processing Methods 0.000 abstract description 4
- 238000013461 design Methods 0.000 abstract description 3
- 238000003672 processing method Methods 0.000 abstract description 3
- 230000000007 visual effect Effects 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 206010019233 Headaches Diseases 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000003491 array Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000002860 competitive effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 231100000869 headache Toxicity 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000009191 jumping Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000013439 planning Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/14—Traffic control systems for road vehicles indicating individual free spaces in parking areas
- G08G1/141—Traffic control systems for road vehicles indicating individual free spaces in parking areas with means giving the indication of available parking spaces
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/04—Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses a parking space management method, which comprises the following steps: acquiring an image of an area where a parking space is located through high-level monitoring; preprocessing the acquired image, extracting and segmenting to extract each effective parking space area; carrying out parking space state detection on the image according to the calculated parking space area and obtaining a parking space state detection result; carrying out parking space state result display and interaction; and finally, the optimal solution of the resources is realized through accurate sensing and reasonable allocation of the parking spaces and the real-time states thereof. Through the deep learning algorithm, the cost of parking space detection is greatly reduced, the efficiency is improved, and the data management of the parking space state can be effectively improved. The further video data processing method, especially the data processing aiming at the training and the runtime reasoning model, perfects the details of the recognition process. Compared with the traditional method, the multi-classification visual design for deep learning not only improves the accuracy of parking space information identification, but also improves the reliability.
Description
Technical Field
The invention relates to the technical field of intelligent parking, in particular to a parking space management method.
Background
The parking problem seems not to be understood by many people, but as large as a city manager, a large property enterprise, as small as a private car owner and a passenger, frequently relates to the problem, according to the statistics of traffic departments, Beijing and Shanghai have millions of gaps, ordinary users can most easily know that the parking spaces of the residential areas and the unit parking spaces are smooth except for the parking spaces of the office and the work, and headache is usually caused by parking in places such as hospitals, schools, business areas of city centers, hot tourist attractions on holidays and the like. And the use problem based on the parking stall has many derived problems, such as the competitive allocation of electric automobile parking stalls and traditional parking stalls, the occupation and the safety of people's activity space, and the like. These problems have been experienced by the society as a whole with many attempts, such as adding parking lots, using time-shared open space short connections to centralize users, etc., although it is more common to have a relatively uniform standard for parking lot management.
At present, a lot of schemes are also provided for managing the parking lot, and the parking lot management can be simply realized by a coarse-grained scheme such as counting the entrance and exit to identify the license plate to complete charging. However, in the long term, the scheme has many problems, such as the potential safety hazard of lack of monitoring, mutually exclusive competition of people and vehicles for the time space of the entrance passageway, vehicle-to-vehicle occupation of vehicles and the like. From the perspective of the car owner and the manager, the order of the parking spaces is often more complex than the statistics, such as the activities of finding the parking spaces, finding the cars, moving the cars, getting the cars and the like, and especially when the parking space resources are short, only charging on time obviously does not solve the problem.
The fine-grained parking lot management has multiple parking space management modes, wherein the fine-grained parking lot management modes comprise a comparison mode, a simple analysis mode and two common schemes of an ultrasonic parking space detector, an indicator light and a mechanical parking space ground lock which are commonly used. The ultrasonic wave scheme is through ultrasonic sensor to parking stall cycle transmission ultrasonic signal, shelters from the signal when the vehicle and then the perception for occuping and show the parking stall state through the pilot lamp, obtains the parking stall state through the pilot lamp that every parking stall extends when the user. The ground lock scheme is to erect and put down through the ground lock and manage whether the parking stall can be parked, and compared with the parking stall, the parking stall management method is more like the use right management of a parking stall. The two schemes are real-time and reliable in parking space state, and relatively easy to realize parking space management, so that the two schemes become management schemes of a lot of parking lots.
Meanwhile, the scheme of ultrasonic waves and ground locks has a plurality of problems, firstly, the data management mode is relatively original, if the internet is further utilized to obtain the parking space state information, a plurality of queuing problems can be solved, and based on the situation or additional equipment or mechanisms are required to independently develop a set of system; secondly, both the parking sensor and the ground lock need to be constructed and wired, and once planning is carried out, adjustment is difficult; it is also considered that the maintenance management of power supply such as access to commercial power or batteries, and batteries, etc. is not a little trouble.
Disclosure of Invention
In view of the above disadvantages, the present invention provides a parking space management method, which greatly reduces the cost of parking space detection and improves the efficiency through a deep learning algorithm, and can effectively improve the data management of the parking space state.
In order to achieve the above purpose, the embodiment of the invention adopts the following technical scheme:
a parking space management method, comprising the steps of:
acquiring an image of an area where a parking space is located;
preprocessing and segmenting the acquired image, extracting and calculating a parking space area;
carrying out parking space state detection on the image according to the calculated parking space area and obtaining a parking space state detection result;
and displaying and interacting the parking space state result.
According to one aspect of the present invention, the acquiring the image of the area where the parking space is located includes: and acquiring a single-frame image of the area where the parking space is located in real time or at regular time through an image acquisition device.
According to one aspect of the present invention, the acquiring the image of the area where the parking space is located includes: the video of the area where the parking space is located is shot in real time through the camera, and a single-frame image is obtained through video decoding.
According to one aspect of the invention, the preprocessing the acquired image to extract and calculate the parking space area comprises the following steps:
carrying out parking space marking segmentation on the image through training;
carrying out perspective transformation on the image after the mark segmentation;
and converting the image after perspective transformation into an image with a specified size through a sampling algorithm.
According to an aspect of the present invention, the detecting the parking space state of the image according to the calculated parking space area and obtaining the parking space state detection result includes:
storing the image data after the preprocessing;
constructing a parking space state classification model based on deep learning;
training a parking space state classification model according to the stored image data;
carrying out parking space state detection on the image through the trained parking space state classification model;
and obtaining and uploading a parking space state detection result.
According to one aspect of the present invention, the detectable parking space state classification types of the parking space state classification model include: the parking space state is formed by combining three parking spaces of parallel, vertical and inclined columns with any parking space state of vehicle presence, vehicle absence and unknown situation.
According to one aspect of the invention, the unknown conditions include reservation, occlusion (not large area), obstruction, damage and other special conditions, and the states of the parking spaces need to be marked separately in the training phase.
According to an aspect of the invention, the saving the pre-processed image data comprises: the preprocessed image data are firstly stored in different marked directories of the same root directory in a file directory distinguishing mode.
According to one aspect of the invention, the pre-processed image data is represented as a numerical matrix of a specified size.
According to one aspect of the invention, the displaying and interacting of the parking space state result comprises the following steps: and analyzing the parking data according to the parking space state result as required.
The implementation of the invention has the advantages that: the parking space management method comprises the following steps: acquiring an image of an area where a parking space is located; preprocessing the acquired image, extracting and calculating a parking space area; carrying out parking space state detection on the image according to the calculated parking space area and obtaining a parking space state detection result; carrying out parking space state result display and interaction; and finally, the optimal solution of the resources is realized through accurate sensing and reasonable allocation of the parking spaces and the real-time states thereof. Through the deep learning algorithm, the cost of parking space detection is greatly reduced, the efficiency is improved, and the data management of the parking space state can be effectively improved. The further video data processing method, especially the data processing aiming at the training and the runtime reasoning model, perfects the details of the recognition process. Compared with the traditional method, the accuracy of parking space information identification is improved and the reliability is also improved through the multi-classification deep learning visual design.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic view illustrating a parking space management method according to the present invention;
fig. 2 is a flowchart illustrating parking space management according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1 and 2, a parking space management method includes the following steps:
step S1: acquiring an image of an area where a parking space is located;
the step S1 of acquiring the image of the area where the parking space is located may specifically be: and acquiring a single-frame image of the area where the parking space is located in real time or at regular time through an image acquisition device. The images of the plurality of image acquisition devices can be acquired in a multi-thread mode.
In practical application, the image acquisition device can be an image acquisition device such as a monitoring camera, a snapshot camera and the like; for example, images are acquired by a high-mounted surveillance camera; the snapshot camera can be set to be a fixed time interval, namely, the photo image of the area where the parking space is located is taken, the time interval can be 0.5 second, 1 second, 2 seconds and the like, and the snapshot camera is set according to actual conditions; the monitoring camera can uninterruptedly shoot the video of the area where the parking space is located in real time for 4 hours. In this embodiment, the monitoring camera is used as an image acquisition device to acquire an image of an area where a parking space is located, for example, an RTS protocol h.264 streaming camera may be used. The streaming video Protocol is that audio and video are required to be divided into small blocks by streaming transmission, the small blocks are sequentially sent and played when being received, a Real-Time streaming Protocol (RTSP) Real-Time streaming Protocol can provide the best quality video for users at any Time, which is common in network cameras, h.264, also called as MPEG-4 part 10, and advanced video coding is a block-oriented video coding standard based on motion compensation, is one of the most common formats for high-precision video recording, compression and distribution, and in practical application, can also be a format such as h.265.
And acquiring a single frame image by decoding the RTSP/H264 code stream in a circulating way. A single-frame image can be obtained after decoding by a video decoder (such as an OpenCV-FFMPEG open source decoder) according to a protocol, a three-dimensional unit8 unsigned 8-bit reshaped Numpy numerical matrix equivalent to a BGR tristimulus image matrix is stored, and the matrix value is an integer from 0 to 255. The method is characterized by comprising the steps of describing by using a camera with the resolution of 720P and 60 frames per second, for example, video data acquired by a video monitoring address RTSP://211.94.164.227/3.3gp RTSP/1.0, and acquiring 60 parts of a numerical matrix with each pixel value of 1280 x 720 x 3 being 0 to 255 through analyzing the video data.
Step S2: preprocessing and segmenting the acquired image, extracting and calculating a parking space area;
from the perspective of the parking space, no matter how the physical environment changes, for example, someone passes through the camera from other places or the illumination changes, as long as the area is not completely shielded or the illumination degree is not extremely low, the area occupied by each parking space can be regarded as the independent standard of the parking space state on the premise that the general camera is not moved. The basic unit of data is a segmented image of each parking area, here interpreted as a geometrically arbitrary convex quadrilateral, represented in the program by 8 bits of a 9-bit 32-reshaping, or 4 two-bit 32-reshaping arrays, respectively.
In this embodiment, the preprocessing, segmenting, extracting, and calculating the parking space region for the acquired image specifically includes: and preprocessing and segmenting the acquired image to extract each effective parking space area.
The step S2 of preprocessing, segmenting, extracting and calculating the parking space region from the acquired image includes:
carrying out parking space marking segmentation on the image through training;
carrying out perspective transformation on the image after the mark segmentation;
and converting the image after perspective transformation into an image with a specified size through a sampling algorithm.
In practical application, a series of basically similar preprocessing is required to be carried out on the image, and three main operations of mark segmentation, perspective conversion and length and width uniform modification are combined. The mark segmentation comprises an automatic segmentation part and a manual adjustment part, wherein the automatic part is realized by retraining a popular Yolo model at present, the input of the model is four vertex point images/characteristics of a parking space image, the condition of high frequency occurrence in adjacent areas is adopted, and finally 4 parking space vertexes are output at minimum and combined into a parking space after the minimum value is subtracted from the sum of any four points. Because the actual effect of the method is not good enough at present, the top of the stop line is not obvious, so that the difference exists between the separation effect of the stop line and the separation effect of manual marking, a manual marking part must be added, the workload of the part is designed to be stored as a file after marking once, and the part can be reused as long as the relative position of the parking space is not changed because the camera does not move. The manual mark is formed by clicking four times clockwise or anticlockwise to form a convex quadrilateral mark.
The perspective transformation is the projection of the map onto a new viewing plane, also called projection mapping. And then converting any convex quadrangle into a square image of 299 by using a sampling algorithm on the intercepted image. The resulting preprocessed image of each parking space is represented as a matrix of values 299 x 3. All parking area images are stored as a numerical matrix, the correlation transformation of which is done as concurrently as possible depending on the hardware throughput.
Step S3: carrying out parking space state detection on the image according to the calculated parking space area and obtaining a parking space state detection result;
the parking space state detection of the image according to the calculated parking space region and the acquisition of the parking space state detection result comprise:
storing the image data after the preprocessing;
constructing a parking space state classification model based on deep learning;
training a parking space state classification model according to the stored image data;
carrying out parking space state detection on the image through the trained parking space state classification model;
and obtaining and uploading a parking space state detection result.
In the training stage, the preprocessed data are firstly stored in different marked directories of the same root directory in a file directory distinguishing mode. The video observation interface has two points, one is that the multi-path camera access needs to merge images and adopt the horizontal and vertical superposition of a matrix, the layout is the interface superposition of integral square times such as 1, 2, 4, 9, 16 and the like with aesthetic support, and all camera addresses are saved through configuration files. And all are normalized to a matrix of a specified size, e.g., 480P, 854 x 480 x 3, and the empty or dropped frame pictures are used to generate an all 0 matrix of the same size. Jumping to a parking space editing page with the current frame as a background through a TAB keyboard, calling, adding, deleting, changing and checking parking space mark model data, realizing the data through polygon filling and connection lines provided by a video processing frame, and overlapping original images by 50% of weight in order to keep transparency.
And the picture data stored in the training stage is imported into the memory again, and data enhancement is performed, wherein the data enhancement comprises 10-degree random rotation, 0.1 random scaling, and 0.1 random length and width offset to generate data and import the data into the model. Due to the adoption of high-performance GPU calculation, 8 thousand images of each batch of data are trained for 30 times, and the model file is saved after the training is finished.
The parking space state classification type detectable by the parking space state classification model comprises the following steps: the parking space state is formed by combining three parking spaces of parallel, vertical and inclined columns with any parking space state of vehicle presence, vehicle absence and unknown situation. The unknown conditions comprise reservation, shielding, obstacles, damage and other special conditions, and the states of all parking spaces need to be marked separately in a training stage. The occlusion is typically a non-large area occlusion.
In practical application, classification types are combined with three types of common parallel, vertical and diagonal parking spaces, 9 states including vehicle, vehicle-free and unknown states are combined respectively, and the unknown states include special conditions such as reservation and shielding. This requires separate labeling during the training phase, which can greatly increase the recognition effect.
The production stage mainly provides the same observation interface, supports the cyclic video recording function, generates the file (such as AVI) by using time as the file name and writing the file (such as XVID) by a universal decoder (such as FFMPEG). The recognition process combines the parking area marked before to intercept the parking space image in real time, and inputs the parking space image after perspective conversion and image size normalization to 299 x 3 so as to train and load the model, and all marked interfaces are adopted in batches. And forming an array through the maximum classification type to be uploaded to a cloud server for distribution, reservation and management.
Step S4: and displaying and interacting the parking space state result.
The parking space state result display and interaction method comprises the following steps: and analyzing the parking data according to the parking space state result as required.
The implementation of the invention has the advantages that: the parking space management method comprises the following steps: acquiring an image of an area where a parking space is located; preprocessing the acquired image, extracting and calculating a parking space area; carrying out parking space state detection on the image according to the calculated parking space area and obtaining a parking space state detection result; carrying out parking space state result display and interaction; and finally, the optimal solution of the resources is realized through accurate sensing and reasonable allocation of the parking spaces and the real-time states thereof. Through the deep learning algorithm, the cost of parking space detection is greatly reduced, the efficiency is improved, and the data management of the parking space state can be effectively improved. The further video data processing method, especially the data processing aiming at the training and the runtime reasoning model, perfects the details of the recognition process. Compared with the traditional method, the accuracy of parking space information identification is improved and the reliability is also improved through the multi-classification deep learning visual design.
The above description is only an embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention disclosed herein are intended to be covered by the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (10)
1. A parking space management method is characterized by comprising the following steps:
acquiring an image of an area where a parking space is located;
preprocessing and segmenting the acquired image, extracting and calculating a parking space area;
carrying out parking space state detection on the image according to the calculated parking space area and obtaining a parking space state detection result;
and displaying and interacting the parking space state result.
2. The parking space management method according to claim 1, wherein the acquiring the image of the area where the parking space is located comprises: and acquiring a single-frame image of the area where the parking space is located in real time or at regular time through an image acquisition device.
3. The parking space management method according to claim 2, wherein the acquiring the image of the area where the parking space is located comprises: the video of the area where the parking space is located is shot in real time through the camera, and a single-frame image is obtained through video decoding.
4. The parking space management method according to claim 1, wherein the preprocessing the acquired image to extract and calculate the parking space area comprises:
carrying out parking space marking segmentation on the image through training;
carrying out perspective transformation on the image after the mark segmentation;
and converting the image after perspective transformation into an image with a specified size through a sampling algorithm.
5. The parking space management method according to claim 4, wherein the detecting the parking space state of the image according to the calculated parking space area and obtaining the parking space state detection result comprises:
storing the image data after the preprocessing;
constructing a parking space state classification model based on deep learning;
training a parking space state classification model according to the stored image data;
carrying out parking space state detection on the image through the trained parking space state classification model;
and obtaining and uploading a parking space state detection result.
6. The parking space management method according to claim 5, wherein the parking space state classification type detectable by the parking space state classification model comprises: the parking space state is formed by combining three parking spaces of parallel, vertical and inclined columns with any parking space state of vehicle presence, vehicle absence and unknown situation.
7. The parking space management method according to claim 6, wherein the unknown conditions include reservation, occlusion, obstruction, damage and other special conditions, and the status of each parking space needs to be marked separately in the training phase.
8. The parking space management method according to claim 5, wherein the saving of the pre-processed image data includes: the preprocessed image data are firstly stored in different marked directories of the same root directory in a file directory distinguishing mode.
9. The parking space management method according to claim 5, wherein the pre-processed image data is represented as a numerical matrix of a specified size.
10. The parking space management method according to any one of claims 1 to 9, wherein the displaying and interacting of the parking space status result comprises the following steps: and analyzing the parking data according to the parking space state result as required.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110999594.1A CN113724527A (en) | 2021-08-29 | 2021-08-29 | Parking space management method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110999594.1A CN113724527A (en) | 2021-08-29 | 2021-08-29 | Parking space management method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113724527A true CN113724527A (en) | 2021-11-30 |
Family
ID=78678685
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110999594.1A Pending CN113724527A (en) | 2021-08-29 | 2021-08-29 | Parking space management method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113724527A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116152719A (en) * | 2023-03-15 | 2023-05-23 | 南通大学 | Parking management method based on high-order video |
| CN117078799A (en) * | 2023-07-31 | 2023-11-17 | 零束科技有限公司 | Special parking space synthesis method and device based on BEV image |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109817013A (en) * | 2018-12-19 | 2019-05-28 | 新大陆数字技术股份有限公司 | Parking stall state identification method and device based on video flowing |
| DE102018222484A1 (en) * | 2018-12-20 | 2020-06-25 | Robert Bosch Gmbh | Method and device for determining an availability state of a parking space |
| CN111476084A (en) * | 2020-02-25 | 2020-07-31 | 福建师范大学 | Deep learning-based parking lot dynamic parking space condition identification method |
-
2021
- 2021-08-29 CN CN202110999594.1A patent/CN113724527A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109817013A (en) * | 2018-12-19 | 2019-05-28 | 新大陆数字技术股份有限公司 | Parking stall state identification method and device based on video flowing |
| DE102018222484A1 (en) * | 2018-12-20 | 2020-06-25 | Robert Bosch Gmbh | Method and device for determining an availability state of a parking space |
| CN111476084A (en) * | 2020-02-25 | 2020-07-31 | 福建师范大学 | Deep learning-based parking lot dynamic parking space condition identification method |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116152719A (en) * | 2023-03-15 | 2023-05-23 | 南通大学 | Parking management method based on high-order video |
| CN117078799A (en) * | 2023-07-31 | 2023-11-17 | 零束科技有限公司 | Special parking space synthesis method and device based on BEV image |
| CN117078799B (en) * | 2023-07-31 | 2024-05-03 | 零束科技有限公司 | Special parking space synthesis method and device based on BEV image |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20230016568A1 (en) | Scenario recreation through object detection and 3d visualization in a multi-sensor environment | |
| Cai et al. | Deep learning-based video system for accurate and real-time parking measurement | |
| WO2023093186A1 (en) | Neural radiation field-based method and apparatus for constructing pedestrian re-identification three-dimensional data set | |
| AU2017204181B2 (en) | Video camera scene translation | |
| Sebe et al. | 3d video surveillance with augmented virtual environments | |
| US10366263B2 (en) | Object detection for video camera self-calibration | |
| CN109598794B (en) | Construction method of three-dimensional GIS dynamic model | |
| US8620026B2 (en) | Video-based detection of multiple object types under varying poses | |
| CN103795976A (en) | Full space-time three-dimensional visualization method | |
| CN113724527A (en) | Parking space management method | |
| Cho et al. | Diml/cvl rgb-d dataset: 2m rgb-d images of natural indoor and outdoor scenes | |
| US20220028114A1 (en) | Method and System for Calibrating a Camera and Localizing Objects Within the Camera Field of View | |
| WO2020211427A1 (en) | Segmentation and recognition method, system, and storage medium based on scanning point cloud data | |
| CN111383204A (en) | Video image fusion method, fusion device, panoramic monitoring system and storage medium | |
| CN105608209A (en) | Video labeling method and video labeling device | |
| CN108471497A (en) | A kind of ship target real-time detection method based on monopod video camera | |
| CN110705366A (en) | Real-time human head detection method based on stair scene | |
| CN116824859A (en) | Intelligent traffic big data analysis system based on Internet of things | |
| Zhang et al. | Longitudinal-scanline-based arterial traffic video analytics with coordinate transformation assisted by 3D infrastructure data | |
| Li et al. | Intelligent transportation video tracking technology based on computer and image processing technology | |
| CN116912517B (en) | Method and device for detecting camera view field boundary | |
| Deng et al. | Deep learning in crowd counting: A survey | |
| Lee et al. | Vehicle counting based on a stereo vision depth maps for parking management | |
| CN111708907A (en) | Target person query method, device, equipment and storage medium | |
| Laureshyn et al. | Automated video analysis as a tool for analysing road user behaviour |
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: 20211130 |
|
| RJ01 | Rejection of invention patent application after publication |