CN111612013A - Parking system based on deep neural network and work flow thereof - Google Patents
Parking system based on deep neural network and work flow thereof Download PDFInfo
- Publication number
- CN111612013A CN111612013A CN201910139507.8A CN201910139507A CN111612013A CN 111612013 A CN111612013 A CN 111612013A CN 201910139507 A CN201910139507 A CN 201910139507A CN 111612013 A CN111612013 A CN 111612013A
- Authority
- CN
- China
- Prior art keywords
- module
- license plate
- neural network
- deep neural
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 66
- 230000011218 segmentation Effects 0.000 claims abstract description 42
- 238000007781 pre-processing Methods 0.000 claims abstract description 37
- 238000000034 method Methods 0.000 claims abstract description 13
- 230000008569 process Effects 0.000 claims abstract description 4
- 238000011176 pooling Methods 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 abstract description 2
- 238000001914 filtration Methods 0.000 description 8
- 239000011159 matrix material Substances 0.000 description 8
- 230000002146 bilateral effect Effects 0.000 description 3
- 238000011161 development Methods 0.000 description 3
- 230000018109 developmental process Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 239000006002 Pepper Substances 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000003672 processing method Methods 0.000 description 2
- 230000003321 amplification Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013145 classification model Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Images
Classifications
-
- 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
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- 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
- G06V10/30—Noise filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V10/507—Summing image-intensity values; Histogram projection analysis
-
- 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
-
- 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
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a parking system based on a deep neural network and a working process thereof, relates to the field of parking, and aims to solve the problem of low management efficiency of the existing parking lot. The license plate segmentation system comprises an image preprocessing module, a database and an auxiliary navigation module, wherein the image preprocessing module is respectively connected with an image acquisition module and a license plate segmentation module, the license plate segmentation module is connected with a deep neural network module, the deep neural network module is connected with the auxiliary navigation module, and the image preprocessing module, the license plate segmentation module, the deep neural network module and the auxiliary navigation module are all connected with the database. The system combines the image algorithm and the artificial intelligence algorithm to realize intelligent management, can be used in the parking lot requiring complex and accurate parking navigation, and effectively improves the management efficiency of the parking lot; the system can also improve the parking navigation problem of the traditional parking lot, can improve the parking efficiency and realize accurate parking for the current society with higher parking demand.
Description
Technical Field
The invention relates to the field of parking, in particular to a deep neural network-based parking system.
Background
With the rapid development of the economic level and the development of science and technology in China, more and more families have private cars, the situation that one family has a plurality of cars is very common, and cars become daily transportation tools for people. The automobile is a non-rail-borne vehicle which is driven by power and provided with 4 wheels or more than 4 wheels, and is mainly used for carrying people and goods and towing the people and goods.
With the development of social economy, the usage amount of automobiles is rapidly increased, and the demand for public parking is increasingly large. In life, public parking lots are mainly composed of open-air parking lots and underground parking lots. Although there are new types of automated parking lots, the occupancy is very small and the cost is very high. Therefore, how to perform efficient automatic management on the conventional parking lot is a problem which needs to be solved urgently.
In a traditional parking lot, a commonly used parking lot management system provides common functions of vehicle statistics, access management, parking charging and the like, but cannot perform fine management, and is lack of subsequent management of vehicles entering the parking lot. Particularly, for large public parking lots such as hospitals and large commercial places, the parking lot management system has the characteristics of high circulation per unit time, large number of parking spaces, complex garage building structure, various passage outlets and the like, a management defect exists in the parking lot due to the fact that a common parking lot management system, when a vehicle owner parks or lifts the vehicle, a large amount of time is wasted due to searching for vacant parking spaces and searching for vehicles, and people are also researching the problem.
Disclosure of Invention
The present invention is directed to a parking system based on a deep neural network, so as to solve the problems mentioned in the background art.
In order to achieve the purpose, the invention provides the following technical scheme:
the utility model provides a parking system based on deep neural network, includes image preprocessing module, license plate segmentation module, deep neural network module, database and supplementary navigation module, image preprocessing module links to each other with image acquisition module and license plate segmentation module respectively, and license plate segmentation module links to each other with deep neural network module, and deep neural network module links to each other with supplementary navigation module, and image preprocessing module, license plate segmentation module, deep neural network module and supplementary navigation module all link to each other with the database.
As a further scheme of the invention: the image acquisition module adopts infrared monocular camera, and it is effectual to make a video recording, can directly adopt the infrared monocular camera in the safety monitoring to carry out the collection of image, need not to increase the camera in addition, reduces equipment cost.
As a further scheme of the invention: the deep neural network module adopts a neural network model module under a TensorFlow framework.
As a further scheme of the invention: the database adopts a MySQL database which is a relational database widely applied at present and has mature technology.
As a further scheme of the invention: the deep neural network module comprises an input layer, a convolutional layer, a pooling layer, a BN layer, a full-connection layer and an activation function layer, the deep neural network module is increasingly applied to computer vision, and has great advantages compared with a traditional algorithm.
The workflow of the parking system based on the deep neural network comprises the following specific steps:
step one, a vehicle enters a parking lot, an image acquisition module acquires license plate information images of the vehicle and the vehicle in the parking lot and sends the license plate information images to an image preprocessing module, the image preprocessing module preprocesses the license plate information images and sends the license plate information images to a license plate segmentation module, the license plate segmentation module performs vertical projection on the preprocessed license plate information images to obtain a histogram of the images, and then performs character segmentation on the histogram to obtain segmentation characters of the license plate;
and step two, the deep neural network module classifies the segmented characters of the license plate to obtain a classification result of each segmented character, license plate number information of the vehicle is obtained and sent to the database, the database gives navigation information and sends the navigation information to the auxiliary navigation module according to map information of the garage, information of the parked vehicle in the garage and the obtained license plate number information of the vehicle, the auxiliary navigation module processes the navigation information to obtain a road map, and the auxiliary navigation module sends the road map to the vehicle.
As a further scheme of the invention: the image preprocessing module comprises the following specific preprocessing steps: the method comprises the steps of converting a license plate information image into a gray image, removing detection interference of image colors, then carrying out histogram equalization, and simultaneously carrying out noise removal, wherein different processing methods are specific to different noises, so that a plurality of denoising methods are used for carrying out image processing, a user-defined low-pass filter is respectively used for removing general noises, a median filter is used for removing salt-pepper noises, and a bilateral filter is used for filtering noises while keeping a clear boundary, after the noises are removed, the image becomes fuzzy and smooth generally, then the gray image is sharpened by using a Laplace method, then the gray image is subjected to sliding convolution, the difference of boundary pixels is improved, and the character segmentation of the license plate at the later stage is facilitated.
Compared with the prior art, the invention has the beneficial effects that:
the system combines the image algorithm and the artificial intelligence algorithm to realize intelligent management, can be used in parking lots including hospitals requiring complex and accurate parking navigation, and effectively improves the management efficiency of the parking lots;
the system can also improve the parking navigation problem of the traditional parking lot, can improve the parking efficiency for the current society with higher parking demand, realizes accurate parking, and has wide application prospect.
Drawings
Fig. 1 is a schematic structural diagram of a deep neural network-based parking system.
Fig. 2 is a flow chart of the work flow of the image preprocessing module in the parking system based on the deep neural network.
FIG. 3 is a flowchart of the operation of the license plate segmentation module in the deep neural network-based parking system.
Fig. 4 is a flow chart of the operation of the deep neural network module in the deep neural network-based parking system.
FIG. 5 is a schematic diagram of a license plate preprocessed by a deep neural network-based parking system.
Fig. 6 is a grayscale histogram obtained by vertical projection of a parking system based on a deep neural network.
Fig. 7 is a convolution flow diagram of a deep neural network based parking system.
Wherein: the system comprises an image acquisition module, an image preprocessing module, a license plate segmentation module, a 4-depth neural network module, a 5-database and a 6-auxiliary navigation module.
Detailed Description
The technical solution of the present patent will be described in further detail with reference to the following embodiments.
Example 1
The utility model provides a parking system based on deep neural network, includes that image preprocessing module 2, license plate divide module 3, deep neural network module 4, database 5 and supplementary navigation module 6, image preprocessing module 2 links to each other with image acquisition module 1 and license plate divide module 3 respectively, and license plate divides module 3 and deep neural network module 4 links to each other, and deep neural network module 4 links to each other with supplementary navigation module 6, and image preprocessing module 2, license plate divide module 3, deep neural network module 4 and supplementary navigation module 6 all link to each other with database 5.
The image acquisition module 1 adopts an infrared monocular camera, the camera shooting effect is good, the infrared monocular camera in the safety monitoring can be directly adopted to collect images, the camera does not need to be additionally arranged, the equipment cost is reduced, and the data stream output by the image acquisition module 1 not only comprises the images, but also comprises the collected camera serial numbers of the image data.
The database 5 adopts a MySQL database which is a relational database 5 widely applied at present, the technology is mature, the database 5 in the system stores preprocessed images, classification information and other data, simultaneously stores geographic information of different parking lots and provides navigation information for the auxiliary navigation module 6.
The workflow of the parking system based on the deep neural network comprises the following specific steps:
step one, a vehicle enters a parking lot, an image acquisition module 1 acquires license plate information images of the vehicle and the vehicle in the parking lot and sends the license plate information images to an image preprocessing module 2, the image preprocessing module 2 grays the license plate information images, a plurality of filtering methods are respectively used for filtering the license plate images to remove image noise, after filtering, in order to improve the edge contrast of the license plate, a Laplacian operator is used for carrying out convolution operation on the images, the images are sharpened, the edge contrast is enhanced, the preprocessed license plate images are sent to a license plate segmentation module 3, the license plate segmentation module 3 carries out vertical projection on the preprocessed license plate information images to obtain a gray statistical histogram of the images, and because license plate characters are standard and uniform and are not adhered, a fixed threshold value is used for carrying out character cutting to obtain segmentation characters of the license plate;
and step two, the deep neural network module 4 classifies the segmentation characters of the license plate to obtain a classification result of each segmentation character, license plate number information of the vehicle is obtained and sent to the database 5, the database 5 gives navigation information according to map information of the garage, information of the parked vehicle in the garage and the obtained license plate number information of the vehicle and sends the navigation information to the auxiliary navigation module 6, the auxiliary navigation module 6 processes the navigation information to obtain a route map, and the auxiliary navigation module 6 sends the route map to the vehicle.
The image preprocessing module 1 comprises the following specific preprocessing steps: the method comprises the steps of converting a license plate information image into a gray image, removing detection interference of image colors, then carrying out histogram equalization, and simultaneously carrying out noise removal, wherein different processing methods are specific to different noises, so that a plurality of denoising methods are used for carrying out image processing, a user-defined low-pass filter is respectively used for removing general noises, a median filter is used for removing salt-pepper noises, and a bilateral filter is used for filtering noises while keeping a clear boundary, after the noises are removed, the image becomes fuzzy and smooth generally, then the gray image is sharpened by using a Laplace method, then the gray image is subjected to sliding convolution, the difference of boundary pixels is improved, and the character segmentation of the license plate at the later stage is facilitated.
Example 2
A parking system based on a deep neural network comprises an image preprocessing module 2, a license plate segmentation module 3, a deep neural network module 4, a database 5 and an auxiliary navigation module 6, wherein the image preprocessing module 2 is respectively connected with an image acquisition module 1 and the license plate segmentation module 3, the license plate segmentation module 3 is connected with the deep neural network module 4, the deep neural network module 4 is connected with the auxiliary navigation module 6, the image preprocessing module 2, the license plate segmentation module 3, the deep neural network module 4 and the auxiliary navigation module 6 are all connected with the database 5, the auxiliary navigation module 6 is a multi-interface module, license plate information and parking lot information are fused to obtain a route map, and the route map is sent to a third party for navigation.
The deep neural network module 4 adopts a neural network model module under a TensorFlow framework.
The deep neural network module 4 comprises a convolution layer, a pooling layer, a BN layer, a full-connection layer and an activation function layer, the application of the type in computer vision is increasingly hot, and compared with the traditional algorithm, the deep neural network module has great advantages, the deep neural network module is a multi-classification model aiming at license plate character classification, and meanwhile, in consideration of the complexity and parameter quantity of the model, small-scale convolution kernels with the sizes of 3 x 3 and 1 x 1 are adopted, fewer network layers are constructed, and the overfitting phenomenon and the calculation load are avoided. In the aspect of model training, only aiming at the classification of Chinese license plate characters, a training set is formed by Chinese character data containing all Chinese provinces for short, 26 capital English letters and 0-9 digital real license plate characters, generally, the more data sets are, the higher the classification precision is, so that data amplification is carried out on the data sets by using data enhancement methods such as random cutting, mirror image and scaling, and the like, a general data set is obtained, and the model obtained by training has good robustness and generalization.
The workflow of the parking system based on the deep neural network comprises the following specific steps:
step one, a vehicle enters a parking lot, an image acquisition module acquires license plate information images of the vehicle and the vehicle in the parking lot and sends the license plate information images to an image preprocessing module, the image preprocessing module preprocesses the license plate information images, the preprocessing comprises histogram equalization, median filtering, bilateral filtering, low-pass filtering and Laplace sharpening, the preprocessed license plate information images are sent to a license plate segmentation module, the license plate segmentation module carries out vertical projection on the preprocessed license plate information images to obtain a histogram of the image, because the size of characters in the license plate is basically consistent and the phenomenon of font overlapping does not exist, the low valley of a histogram fitting curve is just a segmentation point between the characters, and then the histogram is segmented to obtain segmentation characters of the license plate;
and secondly, classifying the segmented characters of the license plate by the deep neural network module to obtain a classification result of each segmented character, obtaining license plate number information of the vehicle and sending the license plate number information to the database, giving navigation information by the database according to the map information of the garage, the information of the parked vehicle in the garage and the obtained license plate number information of the vehicle and sending the navigation information to the auxiliary navigation module, processing the navigation information by the auxiliary navigation module to obtain a road map, sending the road map to the vehicle by the auxiliary navigation module, providing accurate parking navigation information for the vehicle, providing navigation information for searching the vehicle for the owner of the vehicle needing to lift the vehicle, and avoiding the waste of time for searching the parking space and searching the vehicle.
Example 3
The utility model provides a parking system based on deep neural network, includes that image preprocessing module 2, license plate divide module 3, deep neural network module 4, database 5 and supplementary navigation module 6, image preprocessing module 2 links to each other with image acquisition module 1 and license plate divide module 3 respectively, and license plate divides module 3 and deep neural network module 4 links to each other, and deep neural network module 4 links to each other with supplementary navigation module 6, and image preprocessing module 2, license plate divide module 3, deep neural network module 4 and supplementary navigation module 6 all link to each other with database 5.
The parking system based on the deep neural network comprises the following specific steps:
after the parking vehicle enters the parking lot, the auxiliary navigation module displays the map information of the parking lot and the information of the vacant parking spaces through third-party application;
after parking, the image acquisition module 1 acquires a license plate image and transmits the license plate image to the image preprocessing module 2 for image preprocessing, the license plate segmentation module 3 performs character segmentation on the preprocessed license plate image, the deep neural network module 4 classifies characters to obtain a complete license plate number, and data are stored in the database 5;
the database 5 updates the parking space information of the garage; when a user needs to lift a car, after a car owner inputs a license plate number through a third-party application, the database 5 finds a corresponding parking space according to the license plate number, and the user conducts car finding operation through the auxiliary navigation module 6.
The workflow of fig. 4: the model receives a single license plate number character image segmented by a license plate segmentation module as input, a pixel matrix of the image is subjected to sliding convolution operation through a convolution matrix, the matrix after convolution uses an activation function to carry out nonlinear mapping to obtain a characteristic matrix, a pooling layer carries out pooling operation on the characteristic matrix, the size of the matrix is reduced, a high-dimensional characteristic matrix is obtained after passing through a plurality of convolution layers and pooling layers, the characteristic matrix is converted into a characteristic vector, the characteristic vector is input into a full connection layer to carry out linear calculation and nonlinear mapping, finally the characteristic vector is classified by using a softmax classification function to obtain the classification confidence coefficient of the input image, namely the probability that the image is a certain character (Chinese character, letter and number), and the type with the maximum confidence coefficient is taken, namely the license plate character recognized by the module. And after the segmented character images are classified in sequence, the final license plate number is obtained, and the text data of the license plate number is transmitted to the next module.
The workflow of fig. 3: the pixel gray value histogram of the license plate is obtained by vertically projecting the license plate image after passing through the image preprocessing module, because the fonts of the license plate are standard and uniform, and the numbers are not adhered to each other, the waveforms of the gray value are not connected with each other, and the waveform valleys of the character projection are uniform, so that the characters are segmented according to the threshold of the valleys, and meanwhile, the numbers of the license plate have the characteristic of fixed intervals, so that the characters of the license plate can be cut according to the vertically projected histogram and the fixed intervals.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim concerned.
Furthermore, it should be understood that although the present description refers to embodiments, not every embodiment may contain only a single embodiment, and such description is for clarity only, and those skilled in the art should integrate the description, and the embodiments may be combined as appropriate to form other embodiments understood by those skilled in the art.
Claims (7)
1. The utility model provides a parking system based on deep neural network, its characterized in that, includes image preprocessing module (2), license plate segmentation module (3), deep neural network module (4), database (5) and supplementary navigation module (6), image preprocessing module (2) link to each other with image acquisition module (1) and license plate segmentation module (3) respectively, and license plate segmentation module (3) links to each other with deep neural network module (4), and deep neural network module (4) link to each other with supplementary navigation module (6), and image preprocessing module (2), license plate segmentation module (3), deep neural network module (4) and supplementary navigation module (6) all link to each other with database (5).
2. The deep neural network-based parking system according to claim 1, wherein the image acquisition module (1) employs an infrared monocular camera.
3. The deep neural network-based parking system according to claim 1, wherein the deep neural network module (4) employs a neural network model module under a Tensorflow framework.
4. The deep neural network-based parking system according to claim 1, wherein the database (5) employs a MySQL database.
5. The deep neural network-based parking system according to claim 1 or 3, wherein the deep neural network module (4) comprises an input layer, a convolutional layer, a pooling layer, a BN layer, a full-link layer, and an activation function layer.
6. The workflow of the deep neural network based parking system according to any one of claims 1 to 5, comprising the following specific steps:
firstly, a vehicle enters a parking lot, an image acquisition module (1) acquires license plate information images of the vehicle and the vehicle in the parking lot and sends the license plate information images to an image preprocessing module (2), the image preprocessing module (2) preprocesses the license plate information images and sends the license plate information images to a license plate segmentation module (3), the license plate segmentation module (3) vertically projects the preprocessed license plate information images to obtain a histogram of the images, and then character segmentation is carried out on the histogram to obtain segmentation characters of the license plate;
and step two, the deep neural network module (4) classifies the segmentation characters of the license plate to obtain a classification result of each segmentation character, and further obtains license plate number information of the vehicle and sends the license plate number information to the database (5), the database (5) gives navigation information and sends the navigation information to the auxiliary navigation module (6) according to map information of the garage, information of the parked vehicles in the garage and the obtained license plate number information of the vehicle, the auxiliary navigation module (6) processes the navigation information to obtain a road map, and the auxiliary navigation module (6) sends the road map to the vehicle.
7. The workflow of the deep neural network-based parking system according to claim 6, wherein the image preprocessing module (2) comprises the following specific steps: and converting the license plate information image into a gray-scale image, then carrying out histogram equalization, simultaneously carrying out noise removal, sharpening the gray-scale image by using a Laplace method, and then carrying out sliding convolution on the gray-scale image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910139507.8A CN111612013A (en) | 2019-02-26 | 2019-02-26 | Parking system based on deep neural network and work flow thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910139507.8A CN111612013A (en) | 2019-02-26 | 2019-02-26 | Parking system based on deep neural network and work flow thereof |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111612013A true CN111612013A (en) | 2020-09-01 |
Family
ID=72197786
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910139507.8A Pending CN111612013A (en) | 2019-02-26 | 2019-02-26 | Parking system based on deep neural network and work flow thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111612013A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114973691A (en) * | 2022-05-27 | 2022-08-30 | 上海应用技术大学 | Bad driving behavior recording system |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632572A (en) * | 2013-12-17 | 2014-03-12 | 重庆大学 | Intelligent parking method and system |
| CN106251687A (en) * | 2016-08-22 | 2016-12-21 | 重庆交通大学 | A kind of garage intelligent parking management system |
| CN107239778A (en) * | 2017-06-09 | 2017-10-10 | 中国科学技术大学 | The licence plate recognition method of efficiently and accurately |
| CN107545239A (en) * | 2017-07-06 | 2018-01-05 | 南京理工大学 | A kind of deck detection method matched based on Car license recognition with vehicle characteristics |
-
2019
- 2019-02-26 CN CN201910139507.8A patent/CN111612013A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103632572A (en) * | 2013-12-17 | 2014-03-12 | 重庆大学 | Intelligent parking method and system |
| CN106251687A (en) * | 2016-08-22 | 2016-12-21 | 重庆交通大学 | A kind of garage intelligent parking management system |
| CN107239778A (en) * | 2017-06-09 | 2017-10-10 | 中国科学技术大学 | The licence plate recognition method of efficiently and accurately |
| CN107545239A (en) * | 2017-07-06 | 2018-01-05 | 南京理工大学 | A kind of deck detection method matched based on Car license recognition with vehicle characteristics |
Non-Patent Citations (2)
| Title |
|---|
| 李兵 等.: "基于TensorFlow深度学习的车牌识别算法研究", 《成都工业学院学报》 * |
| 窦艳艳等: "基于WiFi定位的地下车库停车系统设计", 《电子科技》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114973691A (en) * | 2022-05-27 | 2022-08-30 | 上海应用技术大学 | Bad driving behavior recording system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110210475B (en) | License plate character image segmentation method based on non-binarization and edge detection | |
| CN110909666B (en) | Night vehicle detection method based on improved YOLOv3 convolutional neural network | |
| CN101872416B (en) | Vehicle license plate recognition method and system of road image | |
| CN111401148A (en) | Road multi-target detection method based on improved multilevel YO L Ov3 | |
| CN104463241A (en) | Vehicle type recognition method in intelligent transportation monitoring system | |
| CN109145900A (en) | A kind of licence plate recognition method based on deep learning | |
| CN106203330A (en) | A kind of vehicle classification method based on convolutional neural networks | |
| CN110866593B (en) | Highway severe weather identification method based on artificial intelligence | |
| CN104715239A (en) | Vehicle color identification method based on defogging processing and weight blocking | |
| CN103824081A (en) | Method for detecting rapid robustness traffic signs on outdoor bad illumination condition | |
| CN108830254B (en) | Fine-grained vehicle type detection and identification method based on data balance strategy and intensive attention network | |
| CN112016556A (en) | Multi-type license plate recognition method | |
| CN115376082B (en) | Lane line detection method integrating traditional feature extraction and deep neural network | |
| CN112819000A (en) | Streetscape image semantic segmentation system, streetscape image semantic segmentation method, electronic equipment and computer readable medium | |
| CN111915583A (en) | Vehicle and pedestrian detection method based on vehicle-mounted thermal infrared imager in complex scene | |
| CN114463684A (en) | Urban highway network-oriented blockage detection method | |
| CN103324958A (en) | License plate locating method based on projection method and SVM under complex background | |
| CN110400469A (en) | Standard based on people's vehicle association analysis, which is driven, is not inconsistent illegal activities method for early warning | |
| CN116597270A (en) | Road damage target detection method based on attention mechanism integrated learning network | |
| CN116469269A (en) | Vehicle information intelligent management method system and method applying visualization technology | |
| CN102184412A (en) | Licence plate number and letter identification method based on minimum-error-rate Bayes classifier | |
| CN111612013A (en) | Parking system based on deep neural network and work flow thereof | |
| CN116229396B (en) | High-speed pavement disease identification and warning method | |
| Zhang et al. | A front vehicle detection algorithm for intelligent vehicle based on improved gabor filter and SVM | |
| CN112733851A (en) | License plate recognition method for optimizing grain warehouse truck based on convolutional neural network |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200901 |