CN111598154A - AI analysis technology-based vehicle identification perception method - Google Patents
AI analysis technology-based vehicle identification perception method Download PDFInfo
- Publication number
- CN111598154A CN111598154A CN202010404815.1A CN202010404815A CN111598154A CN 111598154 A CN111598154 A CN 111598154A CN 202010404815 A CN202010404815 A CN 202010404815A CN 111598154 A CN111598154 A CN 111598154A
- Authority
- CN
- China
- Prior art keywords
- data set
- training
- vehicle
- analysis technology
- perception method
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 21
- 238000005516 engineering process Methods 0.000 title claims abstract description 16
- 230000008447 perception Effects 0.000 title claims abstract description 14
- 238000007781 pre-processing Methods 0.000 claims abstract description 4
- 238000012360 testing method Methods 0.000 claims description 3
- 238000007639 printing Methods 0.000 abstract description 4
- 230000000007 visual effect Effects 0.000 abstract description 2
- 238000013473 artificial intelligence Methods 0.000 description 8
- 238000001514 detection method Methods 0.000 description 3
- 238000013135 deep learning Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a vehicle identification perception method based on an AI analysis technology, which comprises the following steps: constructing a data set; constructing a labeled data set; acquiring the format of the labeled data set, changing a YOLO model according to the labeled data set, and then importing the labeled data set into the YOLO model for training; modifying a weight file called by the YOLO model, and putting a picture to be identified into the YOLO model for preprocessing; the method for recognizing and sensing the vehicle based on the AI analysis technology enables the position of the chromatic sensor to be adjusted more accurately, the adjustment to be more visual and simple, the adjustment efficiency to be improved, the chromatic of a printing system to be more accurate, the color to be more accurate and the printing quality to be higher.
Description
Technical Field
The invention relates to the technical field of AI (Artificial intelligence), in particular to a vehicle identification perception method based on an AI analysis technology.
Background
Hinton, Geoffrey E and Osindero, Simon and Teh, Yee-Whye. A fantleering algorithm for deep belief nets, published in Neural computation,
the deep learning direction is created, and the feasibility of the deep learning method is verified.
Redmon, Joseph, Farhadi, Ali. You only look once: Unifield, real-timeobject detection, published in CVPR,
a YOLO algorithm applied to target detection is provided, and the recognition accuracy and the recognition speed are considered;
girshick Ross, Donahue Jeff, Darrell Trevor, Malik Jittendra Rich Featureachierarchies for access object detection and management segmentation is published in CVPR
An RCNN model is provided, and high accuracy is achieved for object identification;
szegedy Christian, Liu Wei, Jia Yang gang, Sermanet Pierre, Reed Scott, Anguelov Dragomir, Erhan Dumitru, Vanhoucke Vincent, Rabinovich Andrew. Goingdeepper with volumes in CVPR.
The drawbacks of the above technique are as follows:
1. the recognition method of rcon, YOLO can only be used to classify objects in pictures, but the vehicle owner cannot classify the attribute well.
2. There is no vehicle database that fits well with the current application scenario.
3. The blocking and overlapping of sundries can occur in the vehicle identification process.
Disclosure of Invention
The invention aims to provide a vehicle identification and perception method based on an AI analysis technology to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme:
a vehicle recognition perception method based on an AI analysis technology comprises the following steps:
A. constructing a data set;
B. constructing a labeled data set;
C. acquiring the format of the labeled data set, changing a YOLO model according to the labeled data set, and then importing the labeled data set into the YOLO model for training;
D. modifying a weight file called by the YOLO model, and putting a picture to be identified into the YOLO model for preprocessing;
E. and constructing a vehicle length information table according to the vehicle information published on the network and the vehicle types appearing in the scene, and matching the result of the YOLO model training with the information in the table so as to obtain the length information of the vehicle.
As a further technical scheme of the invention: the subject of the data set is a live picture provided by a customer.
As a further technical scheme of the invention: in the step a, in order to prevent overfitting to the current scene, some pictures are searched from the network as the filling of the data set, so that the algorithm can better identify some special scenes.
As a further technical scheme of the invention: in the step A, in order to enable the data set to be in accordance with the training specification, the picture is divided into a training set and a testing set according to the proportion of 7: 3.
As a further technical scheme of the invention: the step B is specifically as follows: and marking the pictures in the data set by using a picture marking tool, and respectively placing a marking result and the original pictures in different folders to ensure that each original picture has a corresponding marking file, thereby constructing a marked data set.
As a further technical scheme of the invention: in step C, since training requires a large amount of computation, the training algorithm needs to be put on the server for computation. And changing the operation command to enable the operation command to be trained by using the GPU on the server, so that the training speed is higher, and obtaining a weight file suitable for the application environment after the training is finished.
As a further technical scheme of the invention: after the processing in step D, the position of the vehicle can be circled in the picture, and the length type of the vehicle can be roughly judged.
Compared with the prior art, the invention has the beneficial effects that: the vehicle identification sensing method based on the AI analysis technology enables the position of the color register sensor to be adjusted more accurately, the adjustment to be more visual and simple, the adjustment efficiency to be improved, the color register of the printing system to be more accurate, the color to be more accurate and the printing quality to be higher.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all 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.
A vehicle recognition perception method based on an AI analysis technology comprises the following steps:
A. and (3) construction of a data set: in order to make the algorithm better suited for the actual application scenario, live pictures provided by the customer are taken as the subject of the data set. In order to prevent overfitting to the current scene, some other pictures are searched from the network as the filling of the data set, so that the algorithm can better identify some special scenes. To fit the data set to the training specification, the pictures were divided into a training set and a test set in a 7:3 ratio.
B. And marking the pictures in the data set by using a picture marking tool, and respectively placing a marking result and the original pictures in different folders to ensure that each original picture has a corresponding marking file, thereby constructing a marked data set.
C. And acquiring the format of the labeled data set, changing the YOLO model according to the labeled data set, and importing the labeled data set into the YOLO model for training. Since training requires a large amount of computation, training algorithms need to be put on the server for computation. And changing the running command to be trained by using the GPU on the server, so that the training speed is higher. And after the training is finished, obtaining a weight file suitable for the application environment.
D. And modifying the weight file called by the YOLO model, and putting the picture to be identified into the YOLO model for preprocessing. After processing, the position of the vehicle can be circled in the picture, and the length type of the vehicle can be roughly judged.
E. And constructing a vehicle length information table according to the vehicle information published on the network and the vehicle types appearing in the scene, and matching the result of the YOLO model training with the information in the table so as to obtain the length information of the vehicle.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof. The present embodiments are therefore to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein.
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 (8)
1. A vehicle recognition perception method based on AI analysis technology is characterized by comprising the following steps:
constructing a data set;
constructing a labeled data set;
acquiring the format of the labeled data set, changing a YOLO model according to the labeled data set, and then importing the labeled data set into the YOLO model for training;
modifying a weight file called by the YOLO model, and putting a picture to be identified into the YOLO model for preprocessing;
and constructing a vehicle length information table according to the vehicle information published on the network and the vehicle types appearing in the scene, and matching the result of the YOLO model training with the information in the table so as to obtain the length information of the vehicle.
2. The AI analysis technology-based vehicle recognition perception method according to claim 1, wherein the subject of the data set is a live picture provided by a customer.
3. The AI analysis technology-based vehicle identification and perception method according to claim 2, wherein in the step a, in order to prevent overfitting to the current scene, some pictures are searched from the network as a filling of data sets, so that the algorithm can better identify some special scenes.
4. The AI analysis technique based vehicle recognition and perception method according to claim 3, wherein in the step A, the pictures are divided into training set and testing set according to 7:3 ratio in order to make the data set conform to the training specification.
5. The AI analysis technology-based vehicle recognition and perception method according to claim 1, wherein the step B specifically includes: and marking the pictures in the data set by using a picture marking tool, and respectively placing a marking result and the original pictures in different folders to ensure that each original picture has a corresponding marking file, thereby constructing a marked data set.
6. The AI analysis technology-based vehicle recognition and perception method according to claim 1, wherein in the step C, since training requires a large amount of computation, a training algorithm needs to be put on a server for computation.
7. And changing the operation command to enable the operation command to be trained by using the GPU on the server, so that the training speed is higher, and obtaining a weight file suitable for the application environment after the training is finished.
8. The AI analysis technology-based vehicle identification and perception method according to claim 1, wherein the position of the vehicle can be circled in the picture after the processing in step D, and the length type of the vehicle can be roughly judged.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010404815.1A CN111598154A (en) | 2020-05-14 | 2020-05-14 | AI analysis technology-based vehicle identification perception method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010404815.1A CN111598154A (en) | 2020-05-14 | 2020-05-14 | AI analysis technology-based vehicle identification perception method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111598154A true CN111598154A (en) | 2020-08-28 |
Family
ID=72183655
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010404815.1A Pending CN111598154A (en) | 2020-05-14 | 2020-05-14 | AI analysis technology-based vehicle identification perception method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111598154A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116756352A (en) * | 2023-06-25 | 2023-09-15 | 北京建科研软件技术有限公司 | Method and device for acquiring construction engineering standard based on AI technology |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109472301A (en) * | 2018-10-26 | 2019-03-15 | 上海新增鼎数据科技有限公司 | A kind of Vehicle length calculation method, device, system and computer equipment |
| CN109829400A (en) * | 2019-01-18 | 2019-05-31 | 青岛大学 | A kind of fast vehicle detection method |
| US20200020121A1 (en) * | 2018-07-13 | 2020-01-16 | Denso International America, Inc. | Dimension estimating system and method for estimating dimension of target vehicle |
| CN110807123A (en) * | 2019-10-29 | 2020-02-18 | 中国科学院上海微***与信息技术研究所 | Vehicle length calculation method, device and system, computer equipment and storage medium |
-
2020
- 2020-05-14 CN CN202010404815.1A patent/CN111598154A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20200020121A1 (en) * | 2018-07-13 | 2020-01-16 | Denso International America, Inc. | Dimension estimating system and method for estimating dimension of target vehicle |
| CN109472301A (en) * | 2018-10-26 | 2019-03-15 | 上海新增鼎数据科技有限公司 | A kind of Vehicle length calculation method, device, system and computer equipment |
| CN109829400A (en) * | 2019-01-18 | 2019-05-31 | 青岛大学 | A kind of fast vehicle detection method |
| CN110807123A (en) * | 2019-10-29 | 2020-02-18 | 中国科学院上海微***与信息技术研究所 | Vehicle length calculation method, device and system, computer equipment and storage medium |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116756352A (en) * | 2023-06-25 | 2023-09-15 | 北京建科研软件技术有限公司 | Method and device for acquiring construction engineering standard based on AI technology |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021073417A1 (en) | Expression generation method and apparatus, device and storage medium | |
| CN111310662B (en) | Flame detection and identification method and system based on integrated deep network | |
| CN106610969A (en) | Multimodal information-based video content auditing system and method | |
| CN106933816A (en) | Across camera lens object retrieval system and method based on global characteristics and local feature | |
| CN110827312B (en) | Learning method based on cooperative visual attention neural network | |
| CN106780639B (en) | Hash coding method based on significance characteristic sparse embedding and extreme learning machine | |
| CN113705290A (en) | Image processing method, image processing device, computer equipment and storage medium | |
| Zhang et al. | No one can escape: A general approach to detect tampered and generated image | |
| CN109255414A (en) | A kind of colour barcode made an inventory for books, books recognition methods, electronic equipment and storage medium | |
| CN111814576A (en) | Shopping receipt picture identification method based on deep learning | |
| CN111222502B (en) | Infrared small target image labeling method and system | |
| CN114333062B (en) | Pedestrian re-recognition model training method based on heterogeneous dual networks and feature consistency | |
| CN112668662B (en) | Outdoor mountain forest environment target detection method based on improved YOLOv3 network | |
| CN114359172A (en) | Cigarette carton multi-face detection and identification method and system under stock or display scene | |
| CN111598154A (en) | AI analysis technology-based vehicle identification perception method | |
| Agarwal et al. | The advent of deep learning-based image forgery detection techniques | |
| CN117437691A (en) | Real-time multi-person abnormal behavior identification method and system based on lightweight network | |
| CN111062393A (en) | Natural scene Chinese character segmentation method based on spectral clustering | |
| CN110852353A (en) | Intersection classification method and equipment | |
| CN116071653A (en) | Automatic extraction method for multi-stage branch structure of tree based on natural image | |
| CN113947568B (en) | Image processing method and device, electronic equipment and storage medium | |
| CN112070048B (en) | Vehicle attribute identification method based on RDSNet | |
| CN112990076A (en) | Data arrangement method and device based on artificial intelligence | |
| CN112560705A (en) | Face detection method and device and electronic equipment | |
| Sulaiman et al. | Image tampering detection using extreme learning machine |
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: 20200828 |
|
| RJ01 | Rejection of invention patent application after publication |