CN107944394A - A kind of video analysis method and system being detected to material in conveyor - Google Patents
A kind of video analysis method and system being detected to material in conveyor Download PDFInfo
- Publication number
- CN107944394A CN107944394A CN201711205692.3A CN201711205692A CN107944394A CN 107944394 A CN107944394 A CN 107944394A CN 201711205692 A CN201711205692 A CN 201711205692A CN 107944394 A CN107944394 A CN 107944394A
- Authority
- CN
- China
- Prior art keywords
- analysis
- video
- digital picture
- frame
- detected
- 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
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
- G06V20/47—Detecting features for summarising video content
-
- 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
- G06V30/19—Recognition using electronic means
- G06V30/192—Recognition using electronic means using simultaneous comparisons or correlations of the image signals with a plurality of references
- G06V30/194—References adjustable by an adaptive method, e.g. learning
Abstract
The invention discloses a kind of video analysis method and system being detected to material in conveyor, available for whether having coal in detection coal conveyor.The coal detection method and system of the present invention is realized in the case where digital camera is as the support of sensor and digital signal processing chip, pass through video acquisition digital video, the digital video of collection is converted into multiframe consecutive digital images, the multiple means such as plane LBP texture Adaboost graders, continuous multiple frames statistical analysis, dynamic LBP texture support vector machine classifiers are merged and are used in combination, finally obtain whether monitoring area has the reliable testing result of material, realize economize on electricity and extend service life of equipment.With higher robustness, there is high practicality.Meet energy-saving, the theory of Green Development.
Description
Technical field
The present invention relates to video detecting method, especially a kind of video analysis method being detected to material in conveyor
And system.
Background technology
Currently, internet+, industry 4.0, artificial intelligence flourish, traditional industry field also begin to largely using intelligence
Robot is artificial to replace, and lifts work efficiency, reduces production cost.Ribbon conveyer is the enterprises such as mine/harbour/power plant
Prevailing traffic equipment, due to the particularity of mining conditions and production link, can not keep ribbon conveyer material freight volume uniform, lead
Cause system or separate unit ribbon conveyer light load or it is non-loaded in the state of run time it is longer, not only cause specified（Or
Setting）The waste of energy run under rotating speed, and belt driving system, rotatable parts, adhesive tape is formed invalid mill
Damage, while shorten service life of equipment.
Ribbon conveyer has largely used frequency conversion drive at this stage, according to transit link condition and freight volume, though periodically
Artificial setting speed governing, but energy-saving effect still unobvious are carried out.In recent years Duo Jia scientific research institutions are energy saving to ribbon conveyer carries out
Substantial amounts of research, freight volume is detected using modes such as batcher quantitative, electronic scale is weighed, infrared rays, and passes through frequency conversion equipment control
Belt speed processed realize it is energy saving, but there is no at present effective, reliable scheme reach in real time detection freight volume change, freight volume distribution, and
Really realize that intelligence adjusts the purpose of belt speed using frequency conversion equipment.In addition, the unlatching of belt type conveyer is needed from last position skin
Band starts successively, general to need more than 50 minutes when the whole start completion of the belt of the entire production line, in the whole process of startup
In, the belt first started is constantly in idling conditions, and the startup of each belt is required for personnel's progress of arranging work.This
Kind Starting mode is because belt idle running causes the waste of electric energy, while increases human cost.Based on video image analysis method
It is a kind of emerging conveyer belt mass flow change detection techniques, which is continuously shot monitoring by the use of video camera as sensor
Region forms live video and is transferred to main control processor, and the intelligent video-image parser operated in main control processor will
The video image of input is handled, is analyzed, identification and statistics, finally detects the real-time change feelings of material on conveyer belt
Condition, and economize on electricity is realized according to the power adjusting variable-frequency motor rotating speed of the rule control frequency converter of setting and extends each component of conveyer belt
The purpose of service life.
The content of the invention
The present invention be it is a kind of based on detection mass flow come be directly realized by regulation motor rotating speed realize economize on electricity method.This
The purpose of invention is to provide that a kind of robustness is more preferable, the material detection method and system based on video analysis of reliability higher.
A kind of material detection method based on video analysis, this method are as sensor and digital signal in video camera
Realized under the support of process chip, this method comprises the following steps：
S1：The simulation or digital camera are continuously shot specified region collection digital video, will collection Video Quality Metric into more
The continuous digital picture of frame；
S2：Described image is detected using the advance trained Adaboost graders based on single frames LBP texture analysis,
Obtain the situation that described image single frames includes mass flow；
S3：S2 steps are repeated, to continuous multiple frames described image analyze simultaneously synthesis result, it is continuously more to obtain the digital picture
Frame includes the situation of mass flow；
S4：S3 steps obtain the situation that described image continuous multiple frames include mass flow, if wherein most of which image bag
Containing material, then monitoring area described in preliminary judgement includes material, otherwise judges that the monitoring area does not include material；
S5：If monitoring area described in preliminary judgement includes material in S4 steps, the branch based on three-dimensional LBP texture analysis is used
Hold vector machine classifier the continuous multiple frames digital picture is further detected and analyzed, according to material on strap surface
Proportion, finally judges how much the monitoring area includes material；
Preferably, the video camera utilizes CCD or CMOS equal modulus conversion chips.
Preferably, the position that the video camera is set up will guarantee to cover the width of coal conveyer belt, and enough faces
Product.
Preferably, it is to utilize the advance trained Adaboost graders based on single frames LBP texture analysis in step S2
The digital picture is detected, obtains the situation that the digital picture single frames includes material.
Preferably, the situation of coal is included in step S4 based on continuous multiple frames analysis, if wherein most of which numeral
Image includes material, then monitoring area described in preliminary judgement includes material, otherwise judges that the monitoring area does not include material feelings
Condition.
Preferably, step S5 uses the support vector machine classifier based on three-dimensional LBP texture analysis to the continuous multiple frames
Digital picture is further detected and analyzed, according to material on strap surface proportion, finally judge the monitored space
How much domain includes material.
A kind of mass flow detecting system based on video analysis, it is used to implement a kind of material inspection based on video analysis
Survey method, it includes video acquisition module：Be continuously shot specified region collection video using video camera, will collection Video Quality Metric into
The continuous digital picture of multiframe；Single-frame analysis module：Single-frame analysis is carried out based on single frames LBP texture Adaboost graders, just
The planar grains of step analysis material, and fast filtering falls the picture frame there is no coal；Multi-frame analysis module：To continuous multiple frames institute
State digital picture to be analyzed, improve the analysis precision of single-frame analysis module, the three-dimensional situation of preliminary analysis material, and quickly
Filter out the picture frame there is no material；Strengthen analysis module：Utilize the support vector cassification based on three-dimensional LBP texture analysis
Device carries out final differentiation, and lifting differentiates credible result degree, strengthens the precision of algorithm.
Preferably, the video camera utilizes CCD or CMOS equal modulus conversion chips.
Preferably, the position that the video camera is set up will guarantee to cover the width of material conveyor belt, and enough faces
Product.
Preferably, it is to utilize the advance trained Adaboost graders based on single frames LBP texture analysis in step S2
The digital picture is detected, obtains the situation that the digital picture single frames includes mass flow.
Preferably, the situation based on continuous multiple frames analysis inclusion material in step S4, if wherein most of which numeral
Image includes material, then monitoring area described in preliminary judgement includes material, otherwise judges that the monitoring area does not include material.
Preferably, step S5 uses the support vector machine classifier based on three-dimensional LBP texture analysis to the continuous multiple frames
Digital picture is further detected and analyzed, according to material on strap surface proportion, finally judge the monitored space
How much domain includes material.
The present invention the material detection method and system based on video analysis, available for detection conveyor on whether have thing
Material.The material detection method and system of the present invention is as under the support of sensor and digital signal processing chip in video camera
Realize, by video acquisition digital video, the digital video of collection is converted into multiframe consecutive digital images, by plane LBP
The multiple means such as texture Adaboost graders, continuous multiple frames statistical analysis, dynamic LBP texture support vector machine classifiers merge
It is used in combination, finally obtains the reliable testing result of monitoring area material real-time traffic information, realizing economize on electricity and extending equipment makes
With the purpose in service life, there is higher robustness, there is high practicality.
Embodiment
It should be noted that in the case where there is no conflict, the feature in embodiment and embodiment in the application can phase
Mutually combination.
Step 1 is to form digital video frequency flow as signal by the use of video camera as sensor continuous acquisition fixed point area image
Source, video camera utilize CCD or CMOS equal modulus conversion chips；The position that video camera is set up will guarantee to cover coal conveyer belt
Width, and enough areas.
It is that then a frame is shot a frame when step 2 video camera shoots coal conveyer belt, i.e., we are first to see that single frames passes
Band image is sent, then single-frame images is by being connected into video image.Therefore, when we prepare to combine single frames plane LBP graders
When carrying out coal detection with two graders of multiframe solid LBP graders, using single frames plane LBP graders as pre-sorting
Device, falls to be entirely free of the picture frame of coal for fast filtering, and using multiframe solid LBP graders as postposition grader, use
In accurately detection mass flow.The present invention uses Adaboost classifier training single frames plane LBP graders as mass flow
The pre-sorting device of detection.
Single frames plane LBP graders of the present invention apply the Adaboost levels based on local binary (LBP) training
Connection grader carrys out the picture frame that selective mechanisms include material.
Adaboost is a kind of iterative algorithm, its core concept is that different graders is trained for same training set
（Weak Classifier）, then these weak classifier sets are got up, form a stronger final classification device（Strong classifier）.It is calculated
Method realizes whether it correct according to the classification of each sample among each training set by varying data distribution in itself,
And the accuracy rate of the general classification of last time, to determine the weights of each sample.The new data set for changing weights is given down
Layer grader is trained, and the grader for finally obtaining each training finally merges, as last Decision Classfication device.
Adaboost training process is as follows：
1. first passing through the study to N number of training sample obtains first Weak Classifier；
2. the sample of misclassification and other new datas are formed into a new N number of training sample together, by this sample
Study obtain second Weak Classifier；
3. the sample of 1 and 2 all misclassifications is added other new samples forms another new N number of training sample, by this
The study of sample obtains the 3rd Weak Classifier；
4. eventually passing through lifting obtains strong classifier.
Step 3 is by after step 2, having obtained the situation that digital picture single frames includes coal.Since material is three-dimensional
, and normal light shine in the same old way under, from appearance or image texture, show very abundant diversity, cause to lead to
Cross single-frame analysis and determine that the precision of coal is inadequate, it is understood that there may be missing inspection or flase drop, and the patent invented by us（Video point
Analyse dedicated video camera）Multiframe statistical analysis is carried out, analytical effect can be lifted, and the image that can extract continuous multiple frames is used
In extraction dynamic LBP features.
It whether there is based on single-frame analysis result judgement coal, in theory failing to judge and judging by accident there are certain probability.In order to
The final high accuracy for differentiating result is improved, present invention utilizes the comprehensive statistics analysis of continuous multiple frames to determine final result,
So as to improve the final confidence level for differentiating result, the accuracy and robustness of algorithm are improved.Step 2 based on continuous multiple frames
Comprehensive statistics analysis method is as follows：
1. the analysis and distinguishing result of each frame is added in result queue；
2. (legal range of F is F if result queue's frame number is less than F>=1 constant, can be adjusted according to the actual requirements),
Then not analyze；
If result queue's frame number is more than F, nearest F frame results are counted（Assuming that F=6 are previously set, if result queue's frame
Number is 10, then rejects above 4 frame numbers, the F frame results for taking 6 last frame numbers i.e. nearest are counted）.
Step 4 is by after step 3, having got video frame images of the continuous multiple frames there may be coal, this step
Advance trained dynamic LBP characteristic support vector machines will be used（SVM）Grader to these image zooming-outs dynamic LBP features simultaneously
Classify, and according to material on strap surface proportion, finally judge the monitoring area include material how much.
Situation comprising coal.
The dynamic LBP feature SVM classifiers that algorithm of the present invention utilizes, not by video frame images as independent
Image carries out separating analysis, but the video that multiple video frame images are formed regards a 3-D solid structure as, using vertical
Body LBP texture blending dynamic texture information, not only considers the texture variations on spatial domain, it is also considered that the texture variations in time domain, increase
The strong identification of textural characteristics.As a result of three-dimensional LBP textural characteristics, thus sample in Distribution center, it is necessary to
Largely the sample with Diversity trains SVM classifier as the learning sample of SVM classifier for collection.SVM classifier is instructed
Practice step：
1. gathering learning sample, learning sample is divided into：Positive sample and negative sample.It is whole that a sample material part carrys out automatic network collection
Reason, a part is from coal excavation operation field shooting.Operation field is excavated in coal, dedicated video camera is set up and coal is transmitted
Band carries out being continuously shot to form video material, by adjusting the erection orientation and shooting angle of video camera, and changes live ring
Border illumination, so as to obtain the abundant coal video material with variation.During post-production sample, by the video of collection
Material resolves into video frame images, and by the image comprising coal and not the image comprising coal separates, and on request will figure
As being made as standard-sized sample image, obtain training the required positive sample of SVM classifier and negative sample.Coal will be included
Sample be divided into positive sample（Classification is denoted as 1）, other samples are divided into negative sample（Classification is denoted as 0）.
2. being learnt using support vector machines to the learning sample with label of offer, final training obtains one
It is a effectively to distinguish comprising coal and the SVM classifier model that does not include two kinds of situations of coal.
3. trained SVM classifier model is with regard to that can be used to, to specifying image classify, export positive class（It is worth for 1）Then
The image for including coal is shown to be, exports negative class（It is worth for 0）Then it is shown to be the image for not including coal.
The present invention uses support vector machines machine learning algorithm model.Support vector machines be used for classification problem be known as support to
Classifier SVCM is measured, realizes that algorithm is known as supporting vector classification SVC algorithms.
When supporting vector classification SVC algorithms realize classification estimation function, there are 3 features：（1）Defined in higher dimensional space
Linear function classify；（2）Classification estimation is realized using global linear minimization；（3）The risk function utilized by
Experience error and a regularization part derived from structural risk minimization form.
Give the data set of a n group relation unknown sample... n, whereinIt is input vector,It is
Desired value, n are the sums of data point.SVM can utilize a Nonlinear Mapping, and data x is mapped to high-dimensional feature space H,
And linear approximation is carried out in this space.From Statistical Learning Theory, which has following form：
（B is constant）（1）
Classification estimation problem is defined as that a loss function is carried out to offer as a tribute the problem of minimizing, and optimal classification function is to pass through
Regularization risk functional is minimized under certain constraints：
（2）
WhereinSo that function is more flat, it is known as regularization term；Section 2 is expected risk functional, can be by different damages
Lose function to determine, constant C>0, penalty, controls the punishment degree to the sample beyond error.Utilize herein- unwise
Feel loss function,
ForIf estimation outputWith desired outputThe absolute value of deviation be less thanWhen, it
Equal to 0；Otherwise, it is equal to outputWith desired outputThe absolute value of deviation subtract, it is non-negative by introducing
Slack variable, then above equation can be exchanged into：
Lagrangian is introduced, we may finally obtain：
Thus, introduce kernel function,It is vectorFeature space () and (
) on inner product, i.e.,.Formula can be changed into above：Can be direct by all computings of kernel function
Calculated in the input space, the selection of kernel function is extremely important to support vector machines.
This is arrived, a conveyer belt material detector has been carried out.Merged plane LBP texture Adaboost graders,
The multiple means such as continuous multiple frames statistical analysis, dynamic LBP texture support vector machine classifiers, which merge, to be used in combination, and greatly improves biography
The accuracy and efficiency with material detector is sent, economize on electricity is realized and extends service life of equipment.With higher robustness, tool
There is high practicality.Meet energy-saving, the theory of Green Development.
Above is the preferable of the present invention is implemented to be illustrated, but the invention is not limited to the implementation
Example, those skilled in the art can also make a variety of equivalent variations on the premise of without prejudice to spirit of the invention or replace
Change, these equivalent deformations or replacement are all contained in the application claim limited range.
Claims (10)
1. a kind of video analysis method being detected to material in conveyor, this method be video camera as sensor and
Realized under the support of process chip, it is characterised in that this method comprises the following steps：
S1：The video camera is continuously shot specified region collection video, will gather Video Quality Metric into the continuous digital picture of multiframe；
S2：The digital picture is carried out using the advance trained Adaboost graders based on single frames LBP texture analysis
Detection, obtains the situation that the digital picture single frames includes material；
S3：S2 steps are repeated, to digital picture described in continuous multiple frames analyze simultaneously synthesis result, the digital picture is obtained and connects
Continuous multiframe includes the situation of material；
S4：S3 steps obtain the situation that the digital picture continuous multiple frames include material, if wherein most of which digitized map
Picture includes material, then monitoring area described in preliminary judgement includes material, otherwise judges that the monitoring area does not include material；
S5：If monitoring area described in preliminary judgement includes material in S4 steps, the branch based on three-dimensional LBP texture analysis is used
Hold vector machine classifier the continuous multiple frames digital picture is further detected and analyzed, finally judge the monitored space
Domain bag material.
A kind of 2. video analysis method being detected to material in conveyor according to claim 1, it is characterised in that
The video camera utilizes CCD modulus conversion chips or CMOS modulus conversion chips.
A kind of 3. video analysis method being detected to material in conveyor according to claim 1, it is characterised in that：
Be in step S2 using the advance trained Adaboost graders based on single frames LBP texture analysis to the digital picture into
Row detection, obtains the situation that the digital picture single frames includes material.
A kind of 4. video analysis method being detected to material in conveyor according to claim 1, it is characterised in that：
Adaboost graders described in step S2 are trained in advance using single frames LBP textural characteristics.
A kind of 5. video analysis method being detected to material in conveyor according to claim 4, it is characterised in that：
S2 steps are repeated in step S3, to digital picture described in continuous multiple frames analyze simultaneously synthesis result, obtains the digital picture
Continuous multiple frames include material.
6. the material detection method according to claim 1 based on video analysis, it is characterised in that：The step S4 includes
Following sub-step：
S41：The analysis and distinguishing result of each frame is added in result queue；
S42：Queue frame number F is set in advance, F may be greater than or the constant equal to 1, if result queue's frame number is less than F,
Not analyze；
S43：If result queue's frame number is more than F, nearest F frame results are counted.
A kind of 7. video analysis method being detected to material in conveyor according to claim 1, it is characterised in that：
Step S5 using the support vector machine classifier based on three-dimensional LBP texture analysis to the continuous multiple frames digital picture into advance one
The detection and analysis of step, finally judge that the monitoring area includes material situation.
A kind of 8. video analysis method being detected to material in conveyor according to claim 7, it is characterised in that：
The step S5 determines whether to continue further to analyze by the use of step 4 analysis result as previous conditional, and further divides
Analysis has used the continuous multiple frames digital picture that S4 steps obtain.
9. a kind of coal detecting system based on video analysis, it is characterised in that the system includes：
Video acquisition module：Multiframe is converted into for being continuously shot specified region collection digital video, and by collection digital video
Continuous digital picture；
Single-frame analysis module：Single-frame analysis, the plane of preliminary analysis coal are carried out based on single frames LBP texture Adaboost graders
Texture, and fast filtering falls the picture frame there is no material；
Multi-frame analysis module：Digital picture described in continuous multiple frames is analyzed, improves the analysis precision of single-frame analysis module,
The three-dimensional situation of preliminary analysis coal, and fast filtering falls the picture frame there is no material；
Strengthen analysis module：Final differentiation, lifting are carried out using the support vector machine classifier based on three-dimensional LBP texture analysis
Differentiate credible result degree, strengthen the precision of algorithm.
10. a kind of material detecting system based on video analysis according to claim 9, it is characterised in that the numeral is taken the photograph
Camera, which utilizes, includes CCD modulus conversion chips or CMOS modulus conversion chips；The single-frame analysis module is utilized based on single
Frame LBP texture Adaboost graders carry out single-frame analysis；The enhancing analysis module is to utilize to be based on solid LBP texture analysis
Support vector machine classifier carry out final differentiation.
Priority Applications (3)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711205692.3A CN107944394A (en) | 2017-11-27 | 2017-11-27 | A kind of video analysis method and system being detected to material in conveyor |
| AU2018201707A AU2018201707A1 (en) | 2017-11-27 | 2018-03-09 | Video analysis method and system for detecting material on conveyor belt |
| ZA2018/01811A ZA201801811B (en) | 2017-11-27 | 2018-03-19 | Video analysis method and system for detecting material on conveyor belt |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711205692.3A CN107944394A (en) | 2017-11-27 | 2017-11-27 | A kind of video analysis method and system being detected to material in conveyor |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107944394A true CN107944394A (en) | 2018-04-20 |
Family
ID=61950129
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711205692.3A Pending CN107944394A (en) | 2017-11-27 | 2017-11-27 | A kind of video analysis method and system being detected to material in conveyor |
Country Status (3)
| Country | Link |
|---|---|
| CN (1) | CN107944394A (en) |
| AU (1) | AU2018201707A1 (en) |
| ZA (1) | ZA201801811B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110675443A (en) * | 2019-09-24 | 2020-01-10 | 西安科技大学 | Coal briquette area detection method for underground coal conveying image |
| CN111723603A (en) * | 2019-03-19 | 2020-09-29 | 杭州海康威视数字技术股份有限公司 | Material monitoring method, system and device |
| CN112040174A (en) * | 2020-07-20 | 2020-12-04 | 西安科技大学 | Underground coal flow visual detection method |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110782436B (en) * | 2019-10-18 | 2023-11-17 | 宁波大学 | Conveyor belt material state detection method based on computer vision |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100926527B1 (en) * | 2008-08-04 | 2009-11-12 | 한국전진기술(주) | Apparatus and method for monitoring and compensating meander of conveyer-belt |
| CN201569481U (en) * | 2010-01-08 | 2010-09-01 | 北京物资学院 | System for dynamically detecting and transmitting materials |
| CN102496001A (en) * | 2011-11-15 | 2012-06-13 | 无锡港湾网络科技有限公司 | Method of video monitor object automatic detection and system thereof |
| CN106228580A (en) * | 2016-07-29 | 2016-12-14 | 李铮 | The detection of a kind of material based on video analysis, power-economizing method and system |
-
2017
- 2017-11-27 CN CN201711205692.3A patent/CN107944394A/en active Pending
-
2018
- 2018-03-09 AU AU2018201707A patent/AU2018201707A1/en not_active Abandoned
- 2018-03-19 ZA ZA2018/01811A patent/ZA201801811B/en unknown
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR100926527B1 (en) * | 2008-08-04 | 2009-11-12 | 한국전진기술(주) | Apparatus and method for monitoring and compensating meander of conveyer-belt |
| CN201569481U (en) * | 2010-01-08 | 2010-09-01 | 北京物资学院 | System for dynamically detecting and transmitting materials |
| CN102496001A (en) * | 2011-11-15 | 2012-06-13 | 无锡港湾网络科技有限公司 | Method of video monitor object automatic detection and system thereof |
| CN106228580A (en) * | 2016-07-29 | 2016-12-14 | 李铮 | The detection of a kind of material based on video analysis, power-economizing method and system |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111723603A (en) * | 2019-03-19 | 2020-09-29 | 杭州海康威视数字技术股份有限公司 | Material monitoring method, system and device |
| CN110675443A (en) * | 2019-09-24 | 2020-01-10 | 西安科技大学 | Coal briquette area detection method for underground coal conveying image |
| CN110675443B (en) * | 2019-09-24 | 2022-12-20 | 西安科技大学 | Coal briquette area detection method for underground coal conveying image |
| CN112040174A (en) * | 2020-07-20 | 2020-12-04 | 西安科技大学 | Underground coal flow visual detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| AU2018201707A1 (en) | 2019-06-13 |
| ZA201801811B (en) | 2019-01-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107944394A (en) | A kind of video analysis method and system being detected to material in conveyor | |
| CN107133943B (en) | A kind of visible detection method of stockbridge damper defects detection | |
| Xu et al. | Fast method of detecting tomatoes in a complex scene for picking robots | |
| CN102495069B (en) | Method for detecting defects of chain belts of zipper on basis of digital image processing | |
| Hu et al. | Multispectral imaging: A new solution for identification of coal and gangue | |
| CN107876429B (en) | Waste non-ferrous metal automatic separation system based on machine vision | |
| CN104574389A (en) | Battery piece chromatism selection control method based on color machine vision | |
| CN108664874A (en) | Underground coal flow rate testing methods based on image recognition | |
| Huang et al. | Detection of human faces using decision trees | |
| CN104056790A (en) | Intelligent potato sorting method and apparatus | |
| CN112893159B (en) | Coal gangue sorting method based on image recognition | |
| CN113820325B (en) | Corn grain direct-harvest impurity-containing rate and breakage rate online detection system and method | |
| CN110163852B (en) | Conveying belt real-time deviation detection method based on lightweight convolutional neural network | |
| CN102169543B (en) | People counting method based on multiple source information fusion and on-line learning | |
| CN104484680A (en) | Multi-model multi-threshold combined pedestrian detection method | |
| CN103646251B (en) | Apple postharvest field classification detection method and system based on embedded technology | |
| CN104574408A (en) | Industry transparent film package detecting method and device based on shape feature extraction | |
| CN108120715A (en) | A kind of corn kernel percentage of damage online test method and device based on machine vision | |
| EP4184448A1 (en) | Method and system for the automatic classification of rocks according to their minerals | |
| CN109284759A (en) | One kind being based on the magic square color identification method of support vector machines (svm) | |
| CN103293167A (en) | Automatic detection method and system of copper grains in foam nickel | |
| CN103927517B (en) | Motion detection method based on human body global feature histogram entropies | |
| CN205538564U (en) | Broken on -line monitoring system of cereal in combine grain tank | |
| CN116416613A (en) | Citrus fruit identification method and system based on improved YOLO v7 | |
| CN116899896A (en) | Red ginseng sorting device and sorting method based on machine vision |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180420 |
|
| WD01 | Invention patent application deemed withdrawn after publication |