CN111009131A - High-order video intelligence parking system based on image recognition - Google Patents
High-order video intelligence parking system based on image recognition Download PDFInfo
- Publication number
- CN111009131A CN111009131A CN201911235191.9A CN201911235191A CN111009131A CN 111009131 A CN111009131 A CN 111009131A CN 201911235191 A CN201911235191 A CN 201911235191A CN 111009131 A CN111009131 A CN 111009131A
- Authority
- CN
- China
- Prior art keywords
- license plate
- vehicle
- parking space
- recognition
- 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.)
- Granted
Links
Images
Classifications
-
- 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
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/24—Aligning, centring, orientation detection or correction of the image
- G06V10/243—Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
-
- 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
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious 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
- G06V20/63—Scene text, e.g. street names
-
- 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
-
- 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
- G06V2201/08—Detecting or categorising vehicles
-
- 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
Abstract
The invention belongs to the technical field of artificial intelligence and image recognition, and particularly relates to a high-order video intelligent parking system based on image recognition. The invention does not depend on real-time vehicle tracking, and realizes the functions of parking state, license plate recognition of the parked vehicle, parking time statistics, whether the parking is standard and the like by carrying out image recognition processing on the periodic camera snapshot images in the technical modes of parking space state detection (vehicle detection and virtual parking space IOU calculation), license plate recognition, forward/backward tracking of the vehicle under the condition of license plate shielding and the like. The image recognition of the invention is operated on the embedded equipment on the tower, parallel resources such as GPU and the like are not needed, and the high-order video intelligent parking system with low cost, low power consumption and high precision is realized depending on a corresponding image recognition algorithm.
Description
Technical Field
The invention belongs to the technical field of artificial intelligence and image recognition, particularly relates to a high-order video intelligent parking system based on image recognition, and more particularly relates to a parking space parking state detection technology, a license plate recognition technology and a vehicle tracking technology based on images, and also comprises an image acquisition method.
Background
The existing high-level video intelligent parking system scheme includes a sensor-based video/image combination mode or a pure video-based image identification mode. If a processing mode of combining the sensor and the video image is adopted, the equipment cost is higher; the existing image recognition mode based on pure video mainly needs to solve the problem of license plate recognition under the condition that the license plate is shielded. The existing pure video-based high-order video intelligent parking system generally judges whether a vehicle has a parking trend according to vehicle track tracking, and if the vehicle has a deceleration (or parking trend), a corresponding camera is triggered to capture an image of a preset position, so that algorithm processing such as vehicle detection, license plate recognition and the like is further performed. The existing pure video-based high-order video intelligent parking system has relatively complex algorithm, relatively low identification precision, generally depends on a parallel computing unit (such as a GPU) and has higher cost.
Disclosure of Invention
The invention aims to realize a front-end unattended high-level video intelligent parking system by a pure video/image-based identification technology. The invention does not depend on real-time vehicle tracking, and realizes the functions of parking state, license plate recognition of the parked vehicle, parking time statistics, whether the parking is standard and the like by carrying out image recognition processing on the periodic camera snapshot images in the technical modes of parking space state detection (vehicle detection and virtual parking space IOU calculation), license plate recognition, forward/backward tracking of the vehicle under the condition of license plate shielding and the like. The image recognition of the invention is operated on the embedded equipment on the tower, parallel resources such as GPU and the like are not needed, and the high-order video intelligent parking system with low cost, low power consumption and high precision is realized depending on a corresponding image recognition algorithm.
In order to achieve the purpose, the invention adopts the following technical scheme:
the high-order video intelligent parking system based on image recognition comprises an image timing acquisition module, a vehicle detection module, a parking space state monitoring module, a license plate recognition module and a vehicle tracking module;
the image timing acquisition module is used for acquiring and storing images of the camera at regular time;
the vehicle detection module is used for detecting vehicles in real time according to the acquired images and outputting vehicle detection results to the parking space state monitoring module;
the parking space state monitoring module is used for monitoring the parking space state in real time according to a vehicle detection result and virtual parking space information, the virtual parking space information is generated after a parking space region is marked and is stored in a server, and the specific method for monitoring the parking space state comprises the following steps: calculating the overlapping rate according to the vehicle detection result and the virtual parking space information, comparing the overlapping rate with a preset threshold, and judging the parking space state according to the comparison result;
the license plate recognition module is used for recognizing license plates according to the acquired images and comprises a license plate detection unit, a license plate inclination estimation and correction unit and a license plate character recognition unit; the license plate detection unit adopts a convolutional neural network to perform feature extraction, then judges whether the acquired picture contains a license plate or not through classification, if so, the picture enters a license plate inclination estimation and correction unit to be processed, otherwise, the license plate identification fails, and a vehicle tracking module is triggered; the license plate inclination estimation and correction unit performs regression calculation to obtain license plate inclination related parameters based on the features extracted by the license plate detection unit, and is used for correcting to obtain a forward license plate picture; the license plate recognition unit extracts features and converts the features to obtain output character string classification according to the obtained forward license plate picture by adopting a deep learning method, processes the character string classification by a CTC algorithm to obtain a license plate recognition result, and outputs the license plate recognition result;
the vehicle tracking module is used for tracking a vehicle track after the license plate recognition fails, specifically, the vehicle tracking module marks the position of a current image of which the license plate recognition fails, tracks the vehicle track by taking the marked position as the center in the stored image, tracks the vehicle entering process and the vehicle exiting process respectively, selects the image from the vehicle entering process and the vehicle exiting process to enter the license plate detection unit again for license plate detection, and if the license plate recognition fails, ends the license plate recognition and uploads the result.
The invention has the beneficial effects that:
(1) high precision: the vehicle detection, the license plate recognition and the vehicle tracking related by the invention are almost recommended to be realized by a deep learning-based method, the condition that the license plate is shielded is fully considered, and the recognition accuracy of the system is improved to the maximum extent in the ways of tracking recognition in the entering process, tracking recognition in the exiting process and the like;
(2) low power consumption: the related image algorithm designed by the invention does not depend on technologies such as real-time vehicle tracking, real-time vehicle detection and the like, but is realized by adopting a method of capturing images and identifying at regular time, and the vehicle tracking technology is started only under the condition that a license plate is not identified, so that the related image identification equipment does not need to be configured with a parallel computing unit (such as a GPU) and has the characteristic of low power consumption.
(3) The cost is low: the image equipment related to the invention is independent of the GPU, so that the hardware cost is relatively low.
Drawings
FIG. 1 is a schematic illustration of the identification process of the present invention;
fig. 2 is a schematic layout of the image recognition apparatus of the present invention.
Detailed Description
The present invention is described in further detail below, while the deployment and implementation of the system are briefly described.
The invention adopts a timing camera to capture images near a parking space, and realizes a high-order video intelligent parking system through technical methods of parking space state detection, inclined license plate recognition, vehicle tracking (forward and reverse), and the like.
The image timing acquisition module can be realized by an image recognition device and is used for capturing images of the camera at regular time and storing the images. The time interval of image capturing is set to N seconds, which can be adjusted according to the real-time requirement of the system, for example, N is set to 1.
The vehicle detection module is realized by adopting an enhanced preprocessing algorithm based on random coherent mixing and a target detection algorithm based on a convolutional neural network.
Wherein the enhanced preprocessing algorithm based on the random coherent mixing can increase the model generalization capability of the vehicle detection algorithm. The concrete implementation steps are as follows:
(1) defining influence factors of a coherent mixing preprocessing algorithm; the influence factors defined by the invention include: the geometric dimension scaling factor, the coherent fusion coefficient, the chroma transformation coefficient, the saturation transformation coefficient and the brightness transformation coefficient of the target frame are used for carrying out randomization generation on the influence factors. The generation method comprises the following steps: the target frame geometric size scaling factor is a uniformly distributed random number in a range of [1/a, a ] (a is a constant, for example, a ═ 4), the coherent fusion coefficient is a uniformly distributed random number in a range of (0,1), the chrominance transform coefficient is a uniformly distributed random number in a range of [ -18,18], and the saturation transform coefficient and the luminance transform coefficient are uniformly distributed random numbers in a range of (0,2/3)
(2) And performing sample enhancement on the vehicle target area image by performing color space change (chromaticity, saturation and brightness) and geometric size scaling on the vehicle area image at one time. The specific implementation steps are as follows: firstly, converting an image of a vehicle target area from an RGB format to an HSV format, and then randomly updating a chroma channel (hue), wherein the updating formula is as follows:
h_new＝h_old+delta_h_rand
wherein h _ old is the chroma channel before updating, delta _ h _ rand is the chroma transform coefficient, and h _ new is the chroma channel after updating. And updating the saturation and the brightness of the image after the chroma updating transformation, wherein the updating formulas are respectively as follows:
satuation_new＝satuation_old*delta_sat_rand
exposure_new＝exposure_old*delta_exp_rand
wherein, the establishment _ old and the exposure _ old are respectively the image saturation and the image brightness of the vehicle target area before updating, the delta _ sat _ rand and the delta _ exp _ rand are respectively a random saturation transformation coefficient and a random brightness transformation coefficient, the establishment _ new and the exposure _ new are respectively the image saturation and the image brightness of the vehicle target area after updating;
finally, the only image converted by image color space transformation (chroma, saturation and brightness) is converted from HSV format to RGB format
(3) And carrying out random coherent mixing on the enhanced vehicle target area image and all other sample images. The specific implementation mode is as follows: firstly, carrying out random scale transformation on the enhanced vehicle target area Image, namely, carrying out Image size scaling on the enhanced vehicle target area Image according to the geometric size scaling factor of the target frame in the step (1) to obtain a randomly scaled vehicle target area Image Resize _ Image. And finally, carrying out random coherent fusion on the Resize _ Image and the images in all the sample Image sets, wherein the formula is as follows:
Image_New＝α*Image_Old+(1-α)*Resize_Image
where α is a random coherent fusion coefficient.
The target detection algorithm based on the convolutional neural network is used for detecting whether a vehicle exists in an image or not and a rectangular position area of the vehicle, and comprises a one-stage target detection algorithm (such as YOLO and SSD), a two-stage target detection algorithm (FasterRCNN and RFCN) and the like. The combination of the target detection algorithm based on the convolutional neural network and the enhanced preprocessing algorithm based on the random coherent mixing can improve the generalization capability of the algorithm model (obtain higher detection accuracy in the actual vehicle detection application process).
The parking space state detection needs to depend on a vehicle detection result and virtual parking space information, and the virtual parking space can be generated in a mode of manually marking a coordinate area and stored in an image recognition server or a rear-end server. The parking space state detection is to calculate the overlapping rate (IOU) according to the vehicle detection result (the position and the area information of the vehicle) and the virtual parking space information; if the overlapping rate exceeds a certain threshold thresh _1, the parking space is considered to be occupied, and if the overlapping rate is lower than the threshold, the parking space is considered to be in an idle state. The calculation formula of the vehicle and parking space overlapping rate (IOU) is as follows:
the license plate recognition is divided into several stages of license plate detection, license plate inclination estimation and correction and license plate character recognition, as shown in fig. 1:
the feature of the license plate detection and the license plate inclination parameter estimation can be extracted by using the same Convolutional Neural Network (CNN), and then classification (whether the license plate is the license plate) and regression (license plate inclination related parameter estimation) operation are carried out.
And the license plate correction can be carried out by inverse transformation of the license plate inclination parameters and a pixel interpolation method to obtain a forward license plate.
The license plate recognition method can be realized by combining deep learning-based feature extraction (such as CNN, RNN and LSTM feature extraction) with CTC classification, and specifically comprises the steps of firstly performing feature extraction on an image obtained by license plate detection and correction through CNN, then performing further feature conversion on the CNN extracted features through RNN or LSTM, and outputting character string classification. And finally, carrying out duplication/null removal processing (deleting repeated and blank character positions) on the character string classification by the CTC to obtain a final license plate recognition result. An Attention mechanism can be added on the basis of a better license plate identification method, and the license plate identification based on the Attention mechanism is to perform weighting processing on the characteristic information of each area after CNN characteristic extraction.
And the vehicle tracking is to track the vehicle when the license plate recognition fails, search the image of which the license plate is not shielded, and then re-perform the license plate recognition.
The vehicle tracking method can adopt a method based on relevant filtering (such as a KCF algorithm) and also can adopt a method combining deep learning feature extraction and RPN (such as a simaese series algorithm).
The vehicle tracking is performed by searching for the vicinity of the area near the position of the history image, starting from the position where the license plate number is not recognized. Specifically, the vehicle-in process tracking (reverse, analysis of the history image from the current image) and the vehicle-out process tracking (forward, analysis from the history image to the current image) can be divided.
And (3) the images obtained after vehicle tracking need to be subjected to license plate recognition again, if the license plate recognition is successful, the license plate recognition is regarded as successful, and the relevant parking space state information and the license plate information are reported to a back-end server. And if the license plate recognition fails, the license plate recognition is considered to fail (vehicle tracking and license plate recognition are not needed), and relevant state and alarm information are reported to a back-end server.
The installation and deployment of the high-order video intelligent parking system need to be considered in combination with the site environment:
1) if more trees are shielded near the parking space, a better implementation method is to install the camera, the tower, the image recognition equipment and other facilities at the corresponding positions opposite to the road, so as to reduce the probability of the license plate being shielded.
2) If no trees are sheltered near the parking space, the camera is installed by adopting an F-shaped pole tower. The height of the cross rods can be adjusted in different fields, and the cross rods with different heights adopt zoom cameras (gunlocks or ball machines) with different focal length ranges to acquire images.
Fig. 2 shows a schematic diagram of a feasible F-shaped pylon deployment, wherein a plurality of "F" -shaped pylons can be installed and covered on a road, and the distance between two F-shaped pylons can be adjusted according to road conditions. As shown in the following figures, 4-8 parking spaces are arranged between the two F-shaped pole towers.
Claims (4)
1. A high-order video intelligent parking system based on image recognition is characterized by comprising an image timing acquisition module, a vehicle detection module, a parking space state monitoring module, a license plate recognition module and a vehicle tracking module;
the image timing acquisition module is used for acquiring and storing images of the camera at regular time;
the vehicle detection module is used for detecting vehicles in real time according to the acquired images and outputting vehicle detection results to the parking space state monitoring module;
the parking space state monitoring module is used for monitoring the parking space state in real time according to a vehicle detection result and virtual parking space information, the virtual parking space information is generated after a parking space region is marked and is stored in a server, and the specific method for monitoring the parking space state comprises the following steps: calculating the overlapping rate according to the vehicle detection result and the virtual parking space information, comparing the overlapping rate with a preset threshold, and judging the parking space state according to the comparison result;
the license plate recognition module is used for recognizing license plates according to the acquired images and comprises a license plate detection unit, a license plate inclination estimation and correction unit and a license plate character recognition unit; the license plate detection unit adopts a convolutional neural network to perform feature extraction, then judges whether the acquired picture contains a license plate or not through classification, if so, the picture enters a license plate inclination estimation and correction unit to be processed, otherwise, the license plate identification fails, and a vehicle tracking module is triggered; the license plate inclination estimation and correction unit performs regression calculation to obtain license plate inclination related parameters based on the features extracted by the license plate detection unit, and is used for correcting to obtain a forward license plate picture; the license plate recognition unit extracts features and converts the features to obtain output character string classification according to the obtained forward license plate picture by adopting a deep learning method, processes the character string classification by a CTC algorithm to obtain a license plate recognition result, and outputs the license plate recognition result;
the vehicle tracking module is used for tracking a vehicle track after the license plate recognition fails, specifically, the vehicle tracking module marks the position of a current image of which the license plate recognition fails, tracks the vehicle track by taking the marked position as the center in the stored image, tracks the vehicle entering process and the vehicle exiting process respectively, selects the image from the vehicle entering process and the vehicle exiting process to enter the license plate detection unit again for license plate detection, and if the license plate recognition fails, ends the license plate recognition and uploads the result.
2. The high-order video intelligent parking system based on image recognition as claimed in claim 1, wherein the image timing acquisition module is used for acquiring camera pictures according to a preset fixed time interval.
3. The high-order video intelligent parking system based on image recognition as claimed in claim 2, wherein the vehicle detection module is used for vehicle detection based on a deep learning target detection method.
4. The high-order video intelligent parking system based on image recognition as claimed in claim 3, wherein the calculation method of the overlapping rate is as follows:
and comparing the overlapping rate with a preset threshold value, if the overlapping rate is greater than the threshold value, judging that the parking space is occupied, and if the overlapping rate is less than the threshold value, judging that the parking space is idle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911235191.9A CN111009131B (en) | 2019-12-05 | 2019-12-05 | High-order video intelligence parking system based on image recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911235191.9A CN111009131B (en) | 2019-12-05 | 2019-12-05 | High-order video intelligence parking system based on image recognition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111009131A true CN111009131A (en) | 2020-04-14 |
| CN111009131B CN111009131B (en) | 2022-04-08 |
Family
ID=70113896
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911235191.9A Active CN111009131B (en) | 2019-12-05 | 2019-12-05 | High-order video intelligence parking system based on image recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111009131B (en) |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112298287A (en) * | 2020-11-13 | 2021-02-02 | 重庆交通大学 | Method for preventing collision of high-speed train groups in cooperative formation under communication interruption condition |
| CN112382104A (en) * | 2020-11-13 | 2021-02-19 | 重庆盘古美天物联网科技有限公司 | Roadside parking management method based on vehicle track analysis |
| CN112541419A (en) * | 2020-12-08 | 2021-03-23 | 北京精英路通科技有限公司 | Image-based positioning method, position relation judging method and device |
| CN112766206A (en) * | 2021-01-28 | 2021-05-07 | 深圳市捷顺科技实业股份有限公司 | High-order video vehicle detection method and device, electronic equipment and storage medium |
| CN112950954A (en) * | 2021-02-24 | 2021-06-11 | 电子科技大学 | Intelligent parking license plate recognition method based on high-position camera |
| CN113705577A (en) * | 2021-04-23 | 2021-11-26 | 中山大学 | License plate recognition method based on deep learning |
| CN113807344A (en) * | 2020-06-11 | 2021-12-17 | 济南市公安局交通警察支队 | Identification method of license plate of muck vehicle and implementation system thereof |
| CN114758511A (en) * | 2022-06-14 | 2022-07-15 | 深圳市城市交通规划设计研究中心股份有限公司 | Sports car overspeed detection system, method, electronic equipment and storage medium |
| CN115022539A (en) * | 2022-05-30 | 2022-09-06 | 深圳云游四海信息科技有限公司 | High-level intelligent parking camera adjusting method and system |
| CN115049993A (en) * | 2022-08-17 | 2022-09-13 | 成都考拉悠然科技有限公司 | Vehicle abnormal stop monitoring method based on deep learning |
| WO2023005808A1 (en) * | 2021-07-29 | 2023-02-02 | 西门子（中国）有限公司 | Method and apparatus for adjusting camera in parking system |
| CN116152719A (en) * | 2023-03-15 | 2023-05-23 | 南通大学 | Parking management method based on high-order video |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103235938A (en) * | 2013-05-03 | 2013-08-07 | 北京国铁华晨通信信息技术有限公司 | Method and system for detecting and identifying license plate |
| CN109635768A (en) * | 2018-12-20 | 2019-04-16 | 深圳市捷顺科技实业股份有限公司 | Parking stall condition detection method, system and relevant device in a kind of picture frame |
| KR20190061543A (en) * | 2017-11-28 | 2019-06-05 | (주)랩투마켓 | Method for arranging freight in shuttle during automatic loading and unloading system |
| CN110135394A (en) * | 2019-05-27 | 2019-08-16 | 杭州亚美利嘉科技有限公司 | Region occupancy and licence plate recognition method and device |
| CN110188749A (en) * | 2019-05-09 | 2019-08-30 | 青岛讯极科技有限公司 | Designated vehicle Vehicle License Plate Recognition System and method under a kind of more vehicles |
-
2019
- 2019-12-05 CN CN201911235191.9A patent/CN111009131B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103235938A (en) * | 2013-05-03 | 2013-08-07 | 北京国铁华晨通信信息技术有限公司 | Method and system for detecting and identifying license plate |
| KR20190061543A (en) * | 2017-11-28 | 2019-06-05 | (주)랩투마켓 | Method for arranging freight in shuttle during automatic loading and unloading system |
| CN109635768A (en) * | 2018-12-20 | 2019-04-16 | 深圳市捷顺科技实业股份有限公司 | Parking stall condition detection method, system and relevant device in a kind of picture frame |
| CN110188749A (en) * | 2019-05-09 | 2019-08-30 | 青岛讯极科技有限公司 | Designated vehicle Vehicle License Plate Recognition System and method under a kind of more vehicles |
| CN110135394A (en) * | 2019-05-27 | 2019-08-16 | 杭州亚美利嘉科技有限公司 | Region occupancy and licence plate recognition method and device |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113807344A (en) * | 2020-06-11 | 2021-12-17 | 济南市公安局交通警察支队 | Identification method of license plate of muck vehicle and implementation system thereof |
| CN112382104A (en) * | 2020-11-13 | 2021-02-19 | 重庆盘古美天物联网科技有限公司 | Roadside parking management method based on vehicle track analysis |
| CN112298287A (en) * | 2020-11-13 | 2021-02-02 | 重庆交通大学 | Method for preventing collision of high-speed train groups in cooperative formation under communication interruption condition |
| CN112298287B (en) * | 2020-11-13 | 2022-04-12 | 重庆交通大学 | Method for preventing collision of high-speed train groups in cooperative formation under communication interruption condition |
| CN112541419A (en) * | 2020-12-08 | 2021-03-23 | 北京精英路通科技有限公司 | Image-based positioning method, position relation judging method and device |
| CN112766206A (en) * | 2021-01-28 | 2021-05-07 | 深圳市捷顺科技实业股份有限公司 | High-order video vehicle detection method and device, electronic equipment and storage medium |
| CN112950954B (en) * | 2021-02-24 | 2022-05-20 | 电子科技大学 | Intelligent parking license plate recognition method based on high-position camera |
| CN112950954A (en) * | 2021-02-24 | 2021-06-11 | 电子科技大学 | Intelligent parking license plate recognition method based on high-position camera |
| CN113705577A (en) * | 2021-04-23 | 2021-11-26 | 中山大学 | License plate recognition method based on deep learning |
| WO2023005808A1 (en) * | 2021-07-29 | 2023-02-02 | 西门子（中国）有限公司 | Method and apparatus for adjusting camera in parking system |
| CN115022539A (en) * | 2022-05-30 | 2022-09-06 | 深圳云游四海信息科技有限公司 | High-level intelligent parking camera adjusting method and system |
| CN114758511A (en) * | 2022-06-14 | 2022-07-15 | 深圳市城市交通规划设计研究中心股份有限公司 | Sports car overspeed detection system, method, electronic equipment and storage medium |
| CN114758511B (en) * | 2022-06-14 | 2022-11-25 | 深圳市城市交通规划设计研究中心股份有限公司 | Sports car overspeed detection system, method, electronic equipment and storage medium |
| CN115049993A (en) * | 2022-08-17 | 2022-09-13 | 成都考拉悠然科技有限公司 | Vehicle abnormal stop monitoring method based on deep learning |
| CN115049993B (en) * | 2022-08-17 | 2022-11-15 | 成都考拉悠然科技有限公司 | Vehicle abnormal stop monitoring method based on deep learning |
| CN116152719A (en) * | 2023-03-15 | 2023-05-23 | 南通大学 | Parking management method based on high-order video |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111009131B (en) | 2022-04-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111009131B (en) | High-order video intelligence parking system based on image recognition | |
| CN105046196B (en) | Front truck information of vehicles structuring output method based on concatenated convolutional neutral net | |
| CN112329776B (en) | License plate detection method and device based on improved CenterNet network | |
| CN111179302B (en) | Moving target detection method and device, storage medium and terminal equipment | |
| CN106991821A (en) | Vehicles peccancy hand-held mobile terminal data collecting system | |
| CN111582118A (en) | Face recognition method and device | |
| CN110222647B (en) | Face in-vivo detection method based on convolutional neural network | |
| CN112528994B (en) | Free angle license plate detection method, license plate recognition method and recognition system | |
| CN108921147B (en) | Black smoke vehicle identification method based on dynamic texture and transform domain space-time characteristics | |
| KR102171384B1 (en) | Object recognition system and method using image correction filter | |
| Lin et al. | Airborne moving vehicle detection for urban traffic surveillance | |
| CN116311205A (en) | License plate recognition method, license plate recognition device, electronic equipment and storage medium | |
| CN111797694A (en) | License plate detection method and device | |
| CN106951831B (en) | Pedestrian detection tracking method based on depth camera | |
| KR101154350B1 (en) | Multiple Images Processing system using Detection of Objects and segmenting - recognizing of High- resolution Image and there of. | |
| CN111985482B (en) | Multi-mode multi-scene automatic license plate recognition method based on deep learning | |
| Wang et al. | Deep learning-based human activity analysis for aerial images | |
| CN112070081B (en) | Intelligent license plate recognition method based on high-definition video | |
| Li et al. | A fog level detection method based on grayscale features | |
| CN113449574A (en) | Method and device for identifying content on target, storage medium and computer equipment | |
| Zhang et al. | Chinese license plate recognition using machine and deep learning models | |
| Yu et al. | Algorithm of license plate localization based on texture analysis | |
| Wang et al. | Multi-feature fusion based region of interest generation method for far-infrared pedestrian detection system | |
| CN106960418A (en) | The algorithm that sleet is removed in video image | |
| Inoue et al. | Automatic recognition of road signs |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |