CN108764278A - A kind of the self study industrial intelligent detecting system and method for view-based access control model - Google Patents
A kind of the self study industrial intelligent detecting system and method for view-based access control model Download PDFInfo
- Publication number
- CN108764278A CN108764278A CN201810336822.5A CN201810336822A CN108764278A CN 108764278 A CN108764278 A CN 108764278A CN 201810336822 A CN201810336822 A CN 201810336822A CN 108764278 A CN108764278 A CN 108764278A
- Authority
- CN
- China
- Prior art keywords
- result
- control unit
- self study
- unit
- industrial intelligent
- 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
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
-
- 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
- G06V2201/06—Recognition of objects for industrial automation
Abstract
The invention discloses the self study industrial intelligent detecting system and method for a kind of view-based access control model, the system comprises：Imaging unit, computing unit, communication unit and control unit.The computing unit includes detector, tracker, Training Control unit, neural network classifier and classifier parameters control unit.The present invention is based on the self study industrial intelligent detecting systems and method of vision, train neural network model to identify the object category detected in image by small lot data；Recognition result is supplied to user as interaction output；User judges to mark as training by manually demarcating verification recognition result, neural metwork training device using this, improves know accuracy rate in real time；Gradually reduce human intervention.This system is suitable for the identification of different objects, and can ceaselessly improve recognition efficiency by advanced optimizing model parameter.
Description
Technical field
The present invention relates to industrial automation, more particularly to the self study industrial intelligent detection system of a kind of view-based access control model
System and method, learn the manual intervention part of industrial flow-line using artificial intelligence and computer vision system, than in full automatically
Amount statistics, quality testing, the identification of material, and assist and gradually substitute the waste of human resources.
Background technology
At present, conventional machines vision using matched method using mainly pattern-recognition is carried out, its shortcoming is that needing needle
To different target developings targetedly algorithmic match, the general algorithm of neither one, however in practical applications, target individual
Ever-changing, traditional computer vision needs a large amount of algorithm optimization to work to realize specific target, for example adds in hardware
Work field, screw just have different models, size and type, if developed for different screws, need largely to develop
Work, this significantly limits the computer vision application of automatic field.
On the other hand, with the development of artificial intelligence, machine learning has prodigious breakthrough, core technology to be to utilize
The classification capacity of deep neural network, can be based on the basis of classification of training data, and the suitable network model of training can be in spy
Determine to be realized in task than artificial higher efficiency and accuracy rate.This weak artificial intelligence technology has answering for bigger in industrial circle
With reason is to execute specific task, without artificial fatigue, will produce the result of stability and high efficiency, result, which has, to be replicated
Property and expansion.
Invention content
Purpose of the present invention is to：The shortcomings and deficiencies for overcoming the above-mentioned prior art provide a kind of self study work of view-based access control model
Industry intelligent checking system and method.
The technical scheme is that：
A kind of self study industrial intelligent detecting system of view-based access control model, the system comprises：
Imaging unit is imaged application scenarios and is converted into digital signal；
Computing unit carries out self study by the input results of imaging unit, and is divided in real time after learning to certain phase
Analysis, and raising study precision is not stopped according to feedback result, analysis result is issued communication unit and control unit；
The analysis result of computing unit is sent to relevant device by communication unit；
Control unit controls relevant device according to analysis result and exports analysis result to outputting and inputting by display port
On interface.
Preferably, the computing unit includes：
The object detected is supplied to tracker by detector in real time;
Tracker will be demarcated the problem of detecting and be supplied to Training Control unit after classifying；
Training Control unit determines that grader needs the parameter changed and position according to classification results；
Neural network classifier to detector result classify and result is passed back to tracking according to newer classifier parameters
Device and output interface；
Classifier parameters control unit, the classifier parameters control unit are defeated according to Training Control unit result and tracker
Go out result update classifier parameters and reaches classification demand.
Preferably, described to output and input interface input picture and output category result.
Preferably, the imaging unit, using CMOS/CCD sensors or optical lens+secondary light source.
Preferably, the computing unit uses but is not limited to CPU/GPU/FPGA/MCU.
Preferably, communication unit is using wirelessly or non-wirelessly network.
A kind of self study industrial intelligent detection method of view-based access control model, including：
1）After imaging unit obtains the image in scene, object detection is carried out according to specific objective, by target to be detected from background
It distinguishes；
2）The study stage, based on manually demarcating as a result, training neural network unit starts to give birth to after obtaining compared with small data sample
At neural network model；
3）Application stage, the neural network model based on generation, real-time analysis result, and in the starting stage, manual feedback is added
Advanced optimize network model；
4）It ultimately generates recognition result and exports to communication unit and control unit.
It is an advantage of the invention that：
The present invention is based on the self study industrial intelligent detecting system and method for vision, on the basis of classification based on training data, instruction
Practice suitable network model, can realize in particular task than artificial higher efficiency and accuracy rate, without artificial fatigue, generate
The result of stability and high efficiency, result have replicability and expansion.
Description of the drawings
The invention will be further described with reference to the accompanying drawings and embodiments：
Fig. 1 is the theory of constitution figure of self study industrial intelligent detecting system of the present invention；
Fig. 2 is the theory of constitution figure of the computing unit of self study industrial intelligent detecting system of the present invention；
Fig. 3 is the flow chart of self study industrial intelligent detection method of the present invention.
Specific implementation mode
In order to make the purpose , technical scheme and advantage of the present invention be clearer, below in conjunction with specific embodiment, to this
Invention is further elaborated.
As shown in Figure 1, the self study industrial intelligent detecting system of the view-based access control model of the present invention, the system comprises：Imaging
Unit, computing unit, communication unit and control unit.Wherein imaging unit includes but not limited to CMOS/CCD sensors, optics
Camera lens and secondary light source, essential core function are to be imaged to application scenarios and be converted into digital signal；The computing unit packet
CPU/GPU/FPGA/MCU is included but be not limited to, is learnt by the input results of imaging unit, and is real after learning to certain phase
Shi Jinhang is analyzed, and does not stop raising study precision according to feedback result and analysis result is issued communication unit and control unit；Institute
It states communication unit and analysis result is sent to by wirelessly or non-wirelessly network by relevant device；Described control unit is according to analysis result
To control relevant device and be outputted results on image by display port.
Specifically, as shown in Fig. 2, the computing unit includes：
The object detected is supplied to tracker by detector in real time;
Tracker will be demarcated the problem of detecting and be supplied to Training Control unit after classifying；
Training Control unit determines that grader needs the parameter changed and position according to classification results；
Neural network classifier to detector result classify and result is passed back to tracking according to newer classifier parameters
Device and output interface；
Classifier parameters control unit, the classifier parameters control unit are defeated according to Training Control unit result and tracker
Go out result update classifier parameters and reaches classification demand.
Output and input interface, input picture and output category result.
A kind of self study industrial intelligent detection method of view-based access control model of the present invention, as shown in figure 3, including：
1）After imaging unit obtains the image in scene, object detection is carried out according to specific objective, by target to be detected from background
It distinguishes；
2）The study stage, based on manually demarcating as a result, training neural network unit starts to give birth to after obtaining compared with small data sample
At neural network model；
3）Application stage, the neural network model based on generation, real-time analysis result, and in the starting stage, manual feedback is added
Advanced optimize network model；
4）It ultimately generates recognition result and exports to communication unit and control unit.
Embodiment
After the self study industrial intelligent detecting system of view-based access control model is deployed in industrial flow-line, carried out for assembly line object
Detection, differentiates assembly line object with background, and detect object by screen display（Such as printing plate, handware etc.）；?
Starting stage will detect that object is classified as normal object, mistake object and defective object by the method manually demarcated,
And object image input neural metwork training device is subjected to model training；After training data reaches minimum data amount, detection system
System is differentiated according to the neural network model after training, and is classified as normal object according to result, mistake object and defective
Object and by result real-time display, while artificial calibration is corrected display result；It is wanted when correction frequency drops to systemic presupposition
After asking, model has and can apply reproducibility.
The above embodiments merely illustrate the technical concept and features of the present invention, and its object is to allow person skilled in the art
It cans understand the content of the present invention and implement it accordingly, it is not intended to limit the scope of the present invention.It is all main according to the present invention
The modification for wanting the Spirit Essence of technical solution to be done, should be covered by the protection scope of the present invention.
Claims (7)
1. a kind of self study industrial intelligent detecting system of view-based access control model, which is characterized in that the system comprises：
Imaging unit is imaged application scenarios and is converted into digital signal；
Computing unit carries out self study by the input results of imaging unit, and is divided in real time after learning to certain phase
Analysis, and raising study precision is not stopped according to feedback result, analysis result is issued communication unit and control unit；
The analysis result of computing unit is sent to relevant device by communication unit；
Control unit controls relevant device according to analysis result and exports analysis result to outputting and inputting by display port
On interface.
2. self study industrial intelligent detecting system according to claim 1, which is characterized in that the computing unit includes：
The object detected is supplied to tracker by detector in real time;
Tracker will be demarcated the problem of detecting and be supplied to Training Control unit after classifying；
Training Control unit determines that grader needs the parameter changed and position according to classification results；
Neural network classifier to detector result classify and result is passed back to tracking according to newer classifier parameters
Device and output interface；
Classifier parameters control unit, the classifier parameters control unit are defeated according to Training Control unit result and tracker
Go out result update classifier parameters and reaches classification demand.
3. self study industrial intelligent detecting system according to claim 2, which is characterized in that described to output and input interface
Input picture and output category result.
4. self study industrial intelligent detecting system according to claim 1, which is characterized in that the imaging unit uses
CMOS/CCD sensors or optical lens+secondary light source.
5. self study industrial intelligent detecting system according to claim 1, which is characterized in that the computing unit use but
It is not limited to CPU/GPU/FPGA/MCU.
6. self study industrial intelligent detecting system according to claim 1, which is characterized in that communication unit using wireless or
Cable network.
7. a kind of self study industrial intelligent detection method of view-based access control model, which is characterized in that including：
1）After imaging unit obtains the image in scene, object detection is carried out according to specific objective, by target to be detected from background
It distinguishes；
2）The study stage, based on manually demarcating as a result, training neural network unit starts to give birth to after obtaining compared with small data sample
At neural network model；
3）Application stage, the neural network model based on generation, real-time analysis result, and in the starting stage, manual feedback is added
Advanced optimize network model；
4）It ultimately generates recognition result and exports to communication unit and control unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810336822.5A CN108764278A (en) | 2018-04-16 | 2018-04-16 | A kind of the self study industrial intelligent detecting system and method for view-based access control model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810336822.5A CN108764278A (en) | 2018-04-16 | 2018-04-16 | A kind of the self study industrial intelligent detecting system and method for view-based access control model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108764278A true CN108764278A (en) | 2018-11-06 |
Family
ID=64010832
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810336822.5A Pending CN108764278A (en) | 2018-04-16 | 2018-04-16 | A kind of the self study industrial intelligent detecting system and method for view-based access control model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108764278A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109754014A (en) * | 2018-12-29 | 2019-05-14 | 北京航天数据股份有限公司 | Industry pattern training method, device, equipment and medium |
| CN109934194A (en) * | 2019-03-20 | 2019-06-25 | 深圳市网心科技有限公司 | Picture classification method, edge device, system and storage medium |
| CN110473306A (en) * | 2019-08-15 | 2019-11-19 | 优估（上海）信息科技有限公司 | A kind of Work attendance method based on face recognition, device and system |
| CN112686183A (en) * | 2021-01-04 | 2021-04-20 | 大陆投资(中国)有限公司 | Remnant detection device, system, method and electronic equipment |
| CN112749811A (en) * | 2020-12-24 | 2021-05-04 | 上海柏珍信息科技有限公司 | Weighing identification system based on non-inductive self-learning AI algorithm |
| CN113083701A (en) * | 2021-03-03 | 2021-07-09 | 浙江博城机器人科技有限公司 | Automatic garbage classification and sorting method based on outdoor garbage image big data analysis |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105930774A (en) * | 2016-04-14 | 2016-09-07 | 中铁大桥勘测设计院集团有限公司 | Automatic bridge bolt come-off identification method based on neural network |
| CN107481231A (en) * | 2017-08-17 | 2017-12-15 | 广东工业大学 | A kind of handware defect classifying identification method based on depth convolutional neural networks |
-
2018
- 2018-04-16 CN CN201810336822.5A patent/CN108764278A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105930774A (en) * | 2016-04-14 | 2016-09-07 | 中铁大桥勘测设计院集团有限公司 | Automatic bridge bolt come-off identification method based on neural network |
| CN107481231A (en) * | 2017-08-17 | 2017-12-15 | 广东工业大学 | A kind of handware defect classifying identification method based on depth convolutional neural networks |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109754014A (en) * | 2018-12-29 | 2019-05-14 | 北京航天数据股份有限公司 | Industry pattern training method, device, equipment and medium |
| CN109754014B (en) * | 2018-12-29 | 2021-04-27 | 北京航天数据股份有限公司 | Industrial model training method, device, equipment and medium |
| CN109934194A (en) * | 2019-03-20 | 2019-06-25 | 深圳市网心科技有限公司 | Picture classification method, edge device, system and storage medium |
| CN110473306A (en) * | 2019-08-15 | 2019-11-19 | 优估（上海）信息科技有限公司 | A kind of Work attendance method based on face recognition, device and system |
| CN112749811A (en) * | 2020-12-24 | 2021-05-04 | 上海柏珍信息科技有限公司 | Weighing identification system based on non-inductive self-learning AI algorithm |
| CN112686183A (en) * | 2021-01-04 | 2021-04-20 | 大陆投资(中国)有限公司 | Remnant detection device, system, method and electronic equipment |
| CN113083701A (en) * | 2021-03-03 | 2021-07-09 | 浙江博城机器人科技有限公司 | Automatic garbage classification and sorting method based on outdoor garbage image big data analysis |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108764278A (en) | A kind of the self study industrial intelligent detecting system and method for view-based access control model | |
| JP7292000B2 (en) | dynamic camera calibration | |
| CN103793719A (en) | Monocular distance-measuring method and system based on human eye positioning | |
| KR102503116B1 (en) | Automatic Data Set Generation Device and their Control Method for AI Machine Learning Learning | |
| WO2022241784A1 (en) | Defect detection method and apparatus, storage medium, and electronic device | |
| CN104143086A (en) | Application technology of portrait comparison to mobile terminal operating system | |
| CN105405115A (en) | Image module group alignment system and method thereof | |
| CN110261069B (en) | Detection method for optical lens | |
| US20220012884A1 (en) | Image analysis system and analysis method | |
| CN106885622A (en) | A kind of big visual field multiple spot three-dimensional vibrating measuring method | |
| CN111259710B (en) | Parking space structure detection model training method adopting parking space frame lines and end points | |
| CN115619787B (en) | UV glue defect detection method, system, equipment and medium | |
| JP6922399B2 (en) | Image processing device, image processing method and image processing program | |
| TWI779957B (en) | Image analysis model establishment method and image analysis apparatus | |
| CN111881740A (en) | Face recognition method, face recognition device, electronic equipment and medium | |
| CN106682604B (en) | Blurred image detection method based on deep learning | |
| WO2021047453A1 (en) | Image quality determination method, apparatus and device | |
| KR102366396B1 (en) | RGB-D Data and Deep Learning Based 3D Instance Segmentation Method and System | |
| CN108289170A (en) | The camera arrangement and method of metering region can be detected | |
| US11393181B2 (en) | Image recognition system and updating method thereof | |
| TWI697846B (en) | Objective identification method and device thereof | |
| FI126481B (en) | Method and system for testing digital imaging devices | |
| CN107798282B (en) | Method and device for detecting human face of living body | |
| US20190014261A1 (en) | Photographing method and photographing system compatible in air and water | |
| TWI747686B (en) | A defect detection method and a defect detection device |
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: 20181106 |
|
| RJ01 | Rejection of invention patent application after publication |