CN109345562A - A kind of traffic picture intelligent dimension system - Google Patents
A kind of traffic picture intelligent dimension system Download PDFInfo
- Publication number
- CN109345562A CN109345562A CN201811118283.4A CN201811118283A CN109345562A CN 109345562 A CN109345562 A CN 109345562A CN 201811118283 A CN201811118283 A CN 201811118283A CN 109345562 A CN109345562 A CN 109345562A
- Authority
- CN
- China
- Prior art keywords
- picture
- traffic
- data
- crossroad
- processing
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/207—Analysis of motion for motion estimation over a hierarchy of resolutions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/20—Analysis of motion
- G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Traffic Control Systems (AREA)
- Closed-Circuit Television Systems (AREA)
Abstract
The invention discloses a kind of traffic picture intelligent dimension systems, including picture processing system, data collection system and data check check system, camera is equipped in the data collection system, camera is located in traffic cross-road, camera produces crossroad monitor video in the process of work, and crossroad monitor video is sent to picture processing system；The picture processing system, which take out to crossroad monitor video, frame and intercepts video pictures, then carries out process of convolution to the video pictures of interception again, then from selected characteristic region on process of convolution treated picture.Invention increases the direction predictions of traffic image, have co-located the state of an object in the picture with the rectangle position and profile coordinate of object；In addition, the picture processing system that the present invention uses can reduce mark process, from the quick mark and intelligent dimension that can be directed to traffic image, realizing, which reduces picture, marks cost, and to mark procedure.
Description
Technical field
The present invention relates to traffic image processing technology, specifically a kind of traffic picture intelligent dimension system.
Background technique
Currently, the mark of picture is the link and artificial intelligence most taken time and effort in the artificial intelligence initial stage
The link that can not be skipped at this stage.By the mark to sample, the training set data of model can be obtained, pass through training set data reality
The training of existing model, it is final to realize intelligence.It is achieved that quickly mark and intelligent dimension will substantially reduce entire artificial intelligence
Cost, also push artificial intelligence fast development.The existing picture mark mostly not no direction prediction of traffic image.
Summary of the invention
The purpose of the present invention is to provide a kind of traffic picture intelligent dimension systems, to solve to propose in above-mentioned background technique
The problem of.
To achieve the above object, the invention provides the following technical scheme:
A kind of traffic picture intelligent dimension system, including picture processing system, data collection system and data check check system,
Camera is equipped in the data collection system, camera is located in traffic cross-road, and camera is given birth in the process of work
Crossroad monitor video is produced, crossroad monitor video is sent to picture processing system；The picture processing system is to cross
Junction surveillance video, which take out, frame and intercepts video pictures, then carries out process of convolution to the video pictures of interception again, then from volume
Selected characteristic region on product processing treated picture, is thereafter again handled the characteristic area picture of selection, and is continued pair
Treated, and picture is labeled processing；The mark processing is labeled to the characteristic area picture of selection, mark knot
Fruit includes the classification of object, the rectangle position coordinate of object, the direction of motion of object, the information such as outline position coordinate of object,
After being labeled to picture, the coordinate data of feature image can be handled, and ultimately form the JSON file set of mark.
As further scheme of the invention: the data check check system is used and is manually checked to JSON file set
Then core carries out model training to the data checked, and optimizes on the basis of model training to model.
As further scheme of the invention: the rectangle position coordinate of the object and the outline position coordinate data of object
The direction of motion of object can be positioned, the direction of motion of object is predicted and marked.
As further scheme of the invention: correcting labeled data after the model optimization and feed back mark processing
Module.
Compared with prior art, invention increases the direction prediction of traffic image, rectangle position and profile with object
Coordinate has co-located the state of an object in the picture；In addition, the picture processing system that the present invention uses can reduce mark
Beam journey, from the quick mark and intelligent dimension that can be directed to traffic image, realizing, which reduces picture, marks cost, and to annotation flow
Cheng Hua.
Detailed description of the invention
Fig. 1 is a kind of block schematic illustration of traffic picture intelligent dimension system.
Specific embodiment
The technical solution of the patent is explained in further detail With reference to embodiment.
Referring to Fig. 1, a kind of traffic picture intelligent dimension system, including picture processing system, data collection system sum number
According to check system is verified, it is equipped with camera in the data collection system, camera is located in traffic cross-road, and camera exists
Crossroad monitor video is produced during work, crossroad monitor video is sent to picture processing system；The picture
Processing system carries out taking out frame and intercepts video pictures to crossroad monitor video, then rolls up again to the video pictures of interception
Product processing, then from selected characteristic region on process of convolution treated picture, the characteristic area picture of selection is carried out again thereafter
Processing, and continue that picture is labeled processing to treated.
Mark processing is labeled to the characteristic area picture of selection, annotation results include object classification,
The rectangle position coordinate of object, the direction of motion of object, the information such as outline position coordinate of object, after being labeled to picture,
The coordinate data of feature image can be handled, the direction of motion of predicted characteristics object be carried out with this, and marked and shown
Come, and ultimately forms the JSON file set of mark；
The rectangle position coordinate of the object and the outline position coordinate data of object can determine the direction of motion of object
Position, to predict the direction of motion of object, and the direction of motion of object is labeled.
Then the data check check system carries out the data checked using manually checking to JSON file set
Model training, and model is optimized on the basis of model training, labeled data is corrected and fed back after model optimization
Processing module is marked, thus to so that the mark to picture is more accurate, to realize intelligent dimension.
Invention increases the direction predictions of traffic image, have co-located one with the rectangle position and profile coordinate of object
The state of a object in the picture.
The picture processing system that the present invention uses can reduce mark process, from the quick mark that can be directed to traffic image
And intelligent dimension, realizing, which reduces picture, marks cost, and to mark procedure.
The preferred embodiment of the patent is described in detail above, but this patent is not limited to above-mentioned embodiment party
Formula within the knowledge of one of ordinary skill in the art can also be under the premise of not departing from this patent objective
Various changes can be made.
Claims (4)
1. a kind of traffic picture intelligent dimension system, including picture processing system, data collection system and data check verification system
System, which is characterized in that camera is equipped in the data collection system, camera is located in traffic cross-road, and camera exists
Crossroad monitor video is produced during work, crossroad monitor video is sent to picture processing system；The picture
Processing system carries out taking out frame and intercepts video pictures to crossroad monitor video, then rolls up again to the video pictures of interception
Product processing, then from selected characteristic region on process of convolution treated picture, the characteristic area picture of selection is carried out again thereafter
Processing, and continue that picture is labeled processing to treated；Mark processing be then to the characteristic area picture of selection into
Rower note, annotation results include the classification of object, the rectangle position coordinate of object, the direction of motion of object, object profile position
The information such as coordinate are set, after being labeled to picture, the coordinate data of feature image can be handled, and ultimately form mark
JSON file set.
2. a kind of traffic picture intelligent dimension system according to claim 1, which is characterized in that the data check verification
Then system carries out model training to the data checked using manually checking to JSON file set, and in model training
On the basis of model is optimized.
3. a kind of traffic picture intelligent dimension system according to claim 1, which is characterized in that the rectangular bit of the object
The outline position coordinate data for setting coordinate and object can position the direction of motion of object, by the direction of motion of object into
Row is predicted and is marked.
4. a kind of traffic picture intelligent dimension system according to claim 2, which is characterized in that will after the model optimization
Labeled data is corrected and feeds back mark processing module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811118283.4A CN109345562A (en) | 2018-09-26 | 2018-09-26 | A kind of traffic picture intelligent dimension system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811118283.4A CN109345562A (en) | 2018-09-26 | 2018-09-26 | A kind of traffic picture intelligent dimension system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109345562A true CN109345562A (en) | 2019-02-15 |
Family
ID=65306351
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811118283.4A Pending CN109345562A (en) | 2018-09-26 | 2018-09-26 | A kind of traffic picture intelligent dimension system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109345562A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111179271A (en) * | 2019-11-22 | 2020-05-19 | 浙江众合科技股份有限公司 | Object angle information labeling method based on retrieval matching and electronic equipment |
| CN111259728A (en) * | 2019-12-20 | 2020-06-09 | 中译语通文娱科技(青岛)有限公司 | Video image information labeling method |
| CN113128382A (en) * | 2021-04-06 | 2021-07-16 | 青岛以萨数据技术有限公司 | Method and system for detecting lane line at traffic intersection |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2001033503A1 (en) * | 1999-11-03 | 2001-05-10 | Cet Technologies Pte Ltd | Image processing techniques for a video based traffic monitoring system and methods therefor |
| CN101398894A (en) * | 2008-06-17 | 2009-04-01 | 浙江师范大学 | Automobile license plate automatic recognition method and implementing device thereof |
| US20130101159A1 (en) * | 2011-10-21 | 2013-04-25 | Qualcomm Incorporated | Image and video based pedestrian traffic estimation |
| CN106127802A (en) * | 2016-06-16 | 2016-11-16 | 南京邮电大学盐城大数据研究院有限公司 | A kind of movement objective orbit method for tracing |
| CN106203433A (en) * | 2016-07-13 | 2016-12-07 | 西安电子科技大学 | In a kind of vehicle monitoring image, car plate position automatically extracts and the method for perspective correction |
| CN106971563A (en) * | 2017-04-01 | 2017-07-21 | 中国科学院深圳先进技术研究院 | Intelligent traffic lamp control method and system |
| CN107134144A (en) * | 2017-04-27 | 2017-09-05 | 武汉理工大学 | A kind of vehicle checking method for traffic monitoring |
-
2018
- 2018-09-26 CN CN201811118283.4A patent/CN109345562A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2001033503A1 (en) * | 1999-11-03 | 2001-05-10 | Cet Technologies Pte Ltd | Image processing techniques for a video based traffic monitoring system and methods therefor |
| CN101398894A (en) * | 2008-06-17 | 2009-04-01 | 浙江师范大学 | Automobile license plate automatic recognition method and implementing device thereof |
| US20130101159A1 (en) * | 2011-10-21 | 2013-04-25 | Qualcomm Incorporated | Image and video based pedestrian traffic estimation |
| CN106127802A (en) * | 2016-06-16 | 2016-11-16 | 南京邮电大学盐城大数据研究院有限公司 | A kind of movement objective orbit method for tracing |
| CN106203433A (en) * | 2016-07-13 | 2016-12-07 | 西安电子科技大学 | In a kind of vehicle monitoring image, car plate position automatically extracts and the method for perspective correction |
| CN106971563A (en) * | 2017-04-01 | 2017-07-21 | 中国科学院深圳先进技术研究院 | Intelligent traffic lamp control method and system |
| CN107134144A (en) * | 2017-04-27 | 2017-09-05 | 武汉理工大学 | A kind of vehicle checking method for traffic monitoring |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111179271A (en) * | 2019-11-22 | 2020-05-19 | 浙江众合科技股份有限公司 | Object angle information labeling method based on retrieval matching and electronic equipment |
| CN111179271B (en) * | 2019-11-22 | 2021-05-11 | 浙江众合科技股份有限公司 | Object angle information labeling method based on retrieval matching and electronic equipment |
| CN111259728A (en) * | 2019-12-20 | 2020-06-09 | 中译语通文娱科技(青岛)有限公司 | Video image information labeling method |
| CN113128382A (en) * | 2021-04-06 | 2021-07-16 | 青岛以萨数据技术有限公司 | Method and system for detecting lane line at traffic intersection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2021042682A1 (en) | Method, apparatus and system for recognizing transformer substation foreign mattter, and electronic device and storage medium | |
| JP6904614B2 (en) | Object detection device, prediction model creation device, object detection method and program | |
| CN109345562A (en) | A kind of traffic picture intelligent dimension system | |
| US11893774B2 (en) | Systems and methods for training machine models with augmented data | |
| US10635936B2 (en) | Method and system for training a neural network to classify objects or events | |
| JP6800351B2 (en) | Methods and devices for detecting burr on electrode sheets | |
| CN111160295B (en) | Video pedestrian re-recognition method based on region guidance and space-time attention | |
| CN102982336A (en) | Method and system for recognition model generation | |
| CN110796201B (en) | Correction method of annotation frame, electronic equipment and storage medium | |
| US20140233802A1 (en) | Increased Quality of Image Objects Based on Depth in Scene | |
| CN104881665A (en) | Chip character identification and verification method and apparatus | |
| CN102706274A (en) | System for accurately positioning mechanical part by machine vision in industrially-structured scene | |
| CN108717789A (en) | A kind of the acquisition mask method and device of vehicle sample | |
| CN112508865A (en) | Unmanned aerial vehicle inspection obstacle avoidance method and device, computer equipment and storage medium | |
| CN109389105A (en) | A kind of iris detection and viewpoint classification method based on multitask | |
| CN110991297A (en) | Target positioning method and system based on scene monitoring | |
| GB2601064A (en) | Artificial image generation for training an object detection system | |
| CN112785557A (en) | Belt material flow detection method and device and belt material flow detection system | |
| CN111091122B (en) | Training and detecting method and device for multi-scale characteristic convolutional neural network | |
| CN108737743B (en) | Video splicing device and video splicing method based on image splicing | |
| WO2003030084A3 (en) | Face recognition from a temporal sequence of face images | |
| TW200606753A (en) | Image processing apparatus and method | |
| CN107369086A (en) | A kind of identity card stamp system and method | |
| Wang et al. | VIMI: Vehicle-infrastructure multi-view intermediate fusion for camera-based 3D object detection | |
| Huang et al. | Enhancing object detection in the dark using U-Net based restoration module |
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: 20190215 |
|
| RJ01 | Rejection of invention patent application after publication |