CN110196152A - The method for diagnosing faults and system of large-scale landscape lamp group based on machine vision - Google Patents
The method for diagnosing faults and system of large-scale landscape lamp group based on machine vision Download PDFInfo
- Publication number
- CN110196152A CN110196152A CN201910247969.1A CN201910247969A CN110196152A CN 110196152 A CN110196152 A CN 110196152A CN 201910247969 A CN201910247969 A CN 201910247969A CN 110196152 A CN110196152 A CN 110196152A
- Authority
- CN
- China
- Prior art keywords
- image
- lamp group
- landscape lamp
- machine vision
- diagnosing faults
- 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
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M11/00—Testing of optical apparatus; Testing structures by optical methods not otherwise provided for
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
Abstract
The method for diagnosing faults and system for the large-scale landscape lamp group based on machine vision that the invention discloses a kind of.This method comprises: obtaining image when landscape lamp group works normally by camera and establishing java standard library;Obtain the image when work of landscape lamp group in real time using camera;Image procossing is carried out to the image of acquisition;It is identified using the images match in match cognization algorithm and java standard library, finds out most like image;And then whether broken down according to the diversity judgement landscape lamp group of realtime graphic and its most like image.The beneficial effects of the present invention are: 1) liberate manpower, the work of observer is reduced, the working condition of landscape lamp group is more accurately obtained.2) staff can be reminded to repair in time, guarantees the normal work of landscape lamp group, beautifies the environment of surrounding.3) whole to be completed by relevant device, so that work intelligence degree further strengthens, substantially increase the degree of automation of work.
Description
Technical field
The present invention relates to the method for diagnosing faults and system of a kind of landscape lamp group, and in particular to a kind of based on machine vision
The method for diagnosing faults and system of large-scale landscape lamp group, belong to field of artificial intelligence.
Background technique
Landscape Lamp is indispensable part in Modern Landscape, and other than ornamental value with higher, there are also illuminate for it
Effect, be chiefly used in the places such as square, public lawn.The presence of Landscape Lamp enriches people's lives, it is therefore desirable to guarantee scape
The normal work for seeing lamp ensures its working condition this requires staff carries out irregular check to Landscape Lamp.To scape
See lamp monitoring work be it is simple uninteresting, need to propose a kind of new technology to change present situation.This technology is removed
Staff can be liberated, moreover it is possible to improve the efficiency and the degree of automation of work.
In order to guarantee the ornamental value of Landscape Lamp, the design of Landscape Lamp shape is all more complicated under normal circumstances, therefore benefit
It is manually monitored and has certain problems.When using being manually monitored: on the one hand because eye-observation ability is limited,
It will appear mistake to see, leak the case where seeing;On the other hand it may also be obtained because of Landscape Lamp configuration design excessively complicated and presence can not be straight
The working condition for connecing lamp group inside the case where checking, such as annular Landscape Lamp is directly to find out, and is needed by other works
Tool, increases the difficulty of work.
In addition to the above both sides reason, it is often more important that staff can not accomplish to monitor in real time and for a long time, work as scape
It sees lamp to break down in staff's rest, staff cannot timely feedback, and influence whether its normal work.But
The problem can be improved by replacing human eye to be monitored Landscape Lamp using camera, accomplish to monitor in real time and for a long time, thus in time
Feedback problem simultaneously reminds staff to repair, and guarantees the normal work of Landscape Lamp.
Summary of the invention
In order to overcome the shortcomings in the prior art, the purpose of the present invention is to provide a kind of large-scale scape based on machine vision
See the method for diagnosing faults and system of lamp group.
To achieve the above object, the present invention adopts the following technical solutions:
The method for diagnosing faults for the landscape lamp group based on machine vision that the present invention provides a kind of, comprising the following steps:
(1) image when landscape lamp group works normally is obtained by camera, carries out image procossing and normalized simultaneously
Establish java standard library;
(2) image when landscape lamp group works is obtained in real time using camera, and carry out image procossing and normalized;
(3) it is identified using the images match in match cognization algorithm and java standard library, finds out most like image;
(4) and then whether broken down according to the diversity judgement landscape lamp group of realtime graphic and standard picture.
Preferably, image when working normally in the step (1) to Landscape Lamp is normalized and mean filter etc.
Relevant criterion library is established in processing.
Preferably, the camera to installation is required to put on 1,2 when the camera more than one used in the step (1)
Equal numbers are grouped the template image in java standard library to camera shooting header laber, and according to the label of camera, find out convenient for the later period
The specific location of the Landscape Lamp of existing failure.The numbers such as 1,2 are put on to camera, obtain different cameras in template library
Label of the image on java standard library also according to camera is marked, and the position of landscape lamp group is obtained convenient for the later period.
Preferably, normalization is all made of in the step (1) and step (2) carrying out processing makes to obtain to obtain image and mark in real time
Image in quasi- library is in the same size.Normalized refers to the length and width for the original image that zooms in or out, and image is made to exist
It is consistent in size, convenient for comparing.
Preferably, Mean Filtering Algorithm reduction external environment is used to bring the dry of noise in the step (1) and step (2)
It disturbs, improves picture quality.
Preferably, handled in the step (1) and step (2) using mean filter, if current pixel point be (a,
B), then select a template, which is made of many pixels adjacent thereto, ask these pixels average value h (a,
B), the gray scale of the pixel and using it selected as before, as follows:
N refers to the number of pixel all in the template in formula, and i refers to i-th in this n pixel
Point, gi(a, b) refers to i-th point of pixel.
If noise N's (a, b) is desired for 0, variance μ, not affected by noise is No (a, b), then noise-containing figure
Picture g (a, b) is as follows after neighborhood averaging:
Preferably, in the step (3) the step of match cognization are as follows:
By the image H obtained in real time handled well be placed on the template image T of java standard library mobile searching and its in size, figure
As identical image in content, region of the image H on image T is called subgraph Tmn, then whole process is exactly to compare figure
As H and subgraph TmnSimilar degree.If being used to measure the matched related coefficient of two images is S, then related coefficient is big
It is small to illustrate image H and subgraph TmnMatching degree, when S maximum, then explanation finds target, and exports image and figure
As the position in java standard library.
Preferably, two images most like obtained in match cognization module are compared in the step (4), is seen
Examine the size of the two difference: if the difference value of the two is less than the threshold value of setting, landscape lamp group is worked normally;If the two
When difference value is greater than the threshold value of setting, then landscape lamp group breaks down.
Preferably, which further includes sounding an alarm that staff is reminded to tie up automatically when landscape lamp group breaks down
The step of repairing.
The fault diagnosis system for the landscape lamp group based on machine vision that the present invention also provides a kind of, for realizing when being executed
The step of method for diagnosing faults of the above-mentioned landscape lamp group based on machine vision, comprising:
Image capture module, the method which is used to execute step (1) and step (2);
Image processing module, the method which is used to execute step (1) and step (2);
Match cognization module, the method which is used to execute step (3);
Breakdown judge module, the method which is used to execute step (4).
Due to the adoption of the above technical scheme, the beneficial effects of the present invention are:
1) real-time monitoring is carried out to lamp group using camera, liberates manpower, reduces the work of observer, more accurately
Acquisition landscape lamp group working condition.
2) fault diagnosis is carried out to lamp group using machine vision, staff can be reminded to repair in time, guarantee landscape
The normal work of lamp group beautifies the environment of surrounding.
3) fault diagnosis is carried out to large-scale landscape lamp group based on machine vision, it is whole to be completed by relevant device, so that work
It is further strengthened as intelligence degree, substantially increases the degree of automation of work.
Detailed description of the invention
Above-mentioned and/or additional aspect of the invention and advantage will become from the description of the embodiment in conjunction with the following figures
Obviously and it is readily appreciated that, in which:
Fig. 1 is the flow chart of the method for diagnosing faults of the large-scale landscape lamp group the present invention is based on machine vision;
Fig. 2 is the flow chart of match cognization module of the present invention.
Specific embodiment
The embodiment of the present invention is described below in detail, examples of the embodiments are shown in the accompanying drawings, wherein from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to
The embodiment of attached drawing description is exemplary, and for explaining only the invention, and is not considered as limiting the invention.
Embodiment 1
As shown in Figure 1, 2, a kind of fault diagnosis method and system of the large-scale landscape lamp group based on machine vision.
The present invention obtains image when landscape lamp group works normally by camera and establishes java standard library;Utilize camera reality
When image when obtaining the work of landscape lamp group, image procossing is carried out to the image of acquisition, utilizes match cognization algorithm and java standard library
In images match identification, find out most like image;And then according to the diversity judgement landscape of realtime graphic and its most like image
Whether lamp group breaks down, and sounds an alarm staff is reminded to repair when a failure occurs.
The present invention is made of four big nucleus modules, respectively image capture module, image processing module, match cognization mould
Block, breakdown judge module.Image capture module is mainly image when acquiring the work of landscape lamp group, and the normal work of acquisition
The image of work builds up java standard library;Image processing module mainly reduces external environment by Mean Filtering Algorithm and brings the dry of noise
It disturbs, improves picture quality;Treated image is mainly found out most phase with the image comparison in java standard library by match cognization module
Like image;Breakdown judge module refers to is to judge landscape lamp group by the difference for comparing two images that a upper module obtains
No failure sounds an alarm automatically when landscape lamp group breaks down and staff is reminded to repair.
The concrete function of each module is as follows:
Module 1: image capture module
According to require installation camera, and using mounted camera to landscape lamp group carry out real-time monitoring, acquisition phase
The image of pass.The image of landscape lamp group when working normally is obtained first, and acquired image is normalized and mean value
Relevant criterion library is established in the processing such as filtering.Then image when landscape lamp group works is obtained in real time using camera, equally to institute
The image of acquisition is normalized, and keeps the image size for being passed to next module and the image in java standard library in the same size.Return
One change handles the length and width for the original image that refers to zooming in or out, and keeps image consistent in size.
The major function of the module is the real-time image for obtaining landscape lamp group, and figure when being worked normally according to landscape lamp group
As establishing java standard library, required when the camera more than one used to camera shooting header laber, and according to the label of camera to mark
Template image label in quasi- library searches the specific location of the landscape lamp group to break down convenient for the later period.
Module 2: image processing module
In image acquisition process, the quality of image is thus main when performing image processing mainly by the interference of noise
Consider removal noise.
The module is mainly handled all images of acquisition, reduces disturbing factor bring in image acquisition process
It influences, to improve the accuracy that machine automatically analyzes.In the present invention, selection mean filter is handled, it refers to choosing
The average value of all pixels in neighborhood of pixels is determined to replace original pixel gray value, if current pixel point is (a, b), the template
It is made of many pixels adjacent thereto, asks the average value h (a, b) of these pixels, and the pixel that it is selected as before
The gray scale of point, as follows:
If noise N's (a, b) is desired for 0, variance μ, not affected by noise is No (a, b), then noise-containing figure
Picture g (a, b) is as follows after neighborhood averaging:
Module 3: match cognization module
The module main function is to have looked for whether in java standard library and the most similar image of acquired image.It will handle well
The image H obtained in real time be placed on the template image T of java standard library mobile searching and it is identical in size, picture material
Region of the image H on image T is called subgraph T by imagemn, then whole process is exactly movement images H and subgraph Tmn
Similar degree.If being used to measure two images matched related coefficients is S, then the size description of related coefficient image H
With subgraph TmnMatching degree, when S maximum, then explanation finds target, and exports image and image in java standard library
Position.
It is as shown in Figure 1 to establish model.
Match cognization module mainly finds the one group image most like with image in template library, and is identified,
The image that will identify that inputs the working condition that next module judges lamp group.
Module 4: breakdown judge module
Breakdown judge module is to sound an alarm prompting for judging whether landscape lamp group breaks down, and when breaking down
Staff repairs.
The module is to compare two images most like in match cognization module, the size of both comparisons difference.
The module set a threshold value, threshold size 5%, if the difference value of the two be less than setting threshold value, landscape lamp group
It works normally;If the difference value of the two is greater than the threshold value of setting, landscape lamp group breaks down, and triggers warning device, reminds
Staff's maintenance and inspection.Staff can determine the scape to break down according to grouping in java standard library and label in maintenance
See the specific location of lamp group.
Overall step of the invention is as shown in Figure 2.
Although an embodiment of the present invention has been shown and described, it will be understood by those skilled in the art that: not
A variety of change, modification, replacement and modification can be carried out to these embodiments in the case where being detached from the principle of the present invention and objective,
The scope of the present invention is defined by the claims and their equivalents.
Claims (9)
1. a kind of method for diagnosing faults of the landscape lamp group based on machine vision, which comprises the following steps:
(1) image when landscape lamp group works normally is obtained by camera, carries out image procossing and normalized and establishes
Java standard library;
(2) image when landscape lamp group works is obtained in real time using camera, and carry out image procossing and normalized;
(3) it is identified using the images match in match cognization algorithm and java standard library, finds out most like image;
(4) and then whether broken down according to the diversity judgement landscape lamp group of realtime graphic and its most like image.
2. the method for diagnosing faults of the landscape lamp group according to claim 1 based on machine vision, which is characterized in that described
It is required when the camera more than one used in step (1) to camera shooting header laber, and according to the label of camera to java standard library
In template image label, convenient for the later period search break down landscape lamp group specific location.
3. the method for diagnosing faults of the landscape lamp group according to claim 1 based on machine vision, which is characterized in that described
Image when working normally in step (1) to Landscape Lamp carries out mean filter and normalized establishes relevant criterion library.
4. the method for diagnosing faults of the landscape lamp group according to claim 1 based on machine vision, which is characterized in that described
Mean filter is all made of in step (1) and step (2) to be handled, if current pixel point is (a, b), then selectes a mould
Plate, the template are made of many pixels adjacent thereto, seek the average value h (a, b) of these pixels, and it is selected as before
The gray scale for the pixel selected, as follows:
If noise N's (a, b) is desired for 0, variance μ, not affected by noise is No (a, b), then noise-containing image g
(a, b) is as follows after neighborhood averaging:
5. the method for diagnosing faults of the landscape lamp group according to claim 1 based on machine vision, which is characterized in that described
Normalization is all made of in step (1) and step (2) to carry out handling the image size one in the image and java standard library for making to obtain in real time
It causes;Normalized refers to the length and width for the original image that zooms in or out, and keeps image consistent in size, convenient for than
Compared with.
6. the method for diagnosing faults of the landscape lamp group according to claim 1 based on machine vision, which is characterized in that described
In step (3) the step of match cognization are as follows:
By the image H obtained in real time handled well be placed on the template image T of java standard library mobile searching and its in size, image
Region of the image H on image T is called subgraph T by identical image in appearancemn, then whole process be exactly movement images H and
Subgraph TmnSimilar degree;If being used to measure the matched related coefficient of two images is S, then the size description of related coefficient
Image H and subgraph TmnMatching degree, when S maximum, then explanation finds target, and exports image and image in standard
Position in library.
7. the method for diagnosing faults of the landscape lamp group according to claim 1 based on machine vision, which is characterized in that described
Two images most like obtained in match cognization module are compared in step (4), the size of both observations difference: if
When the difference value of the two is less than the threshold value of setting, then landscape lamp group works normally;If the difference value of the two is greater than the threshold value of setting
When, then landscape lamp group breaks down.
8. the method for diagnosing faults of the landscape lamp group according to claim 1 based on machine vision, which is characterized in that the step
Suddenly further including the steps that sounding an alarm automatically when landscape lamp group breaks down reminds staff to repair.
9. a kind of fault diagnosis system of the landscape lamp group based on machine vision, which is characterized in that for realizing power when being executed
The step of method for diagnosing faults of the benefit landscape lamp group based on machine vision that requires 1-8 described in any item, comprising:
Image capture module, the method which is used to execute step (1) and step (2);
Image processing module, the method which is used to execute step (1) and step (2);
Match cognization module, the method which is used to execute step (3);
Breakdown judge module, the method which is used to execute step (4).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910247969.1A CN110196152A (en) | 2019-03-29 | 2019-03-29 | The method for diagnosing faults and system of large-scale landscape lamp group based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910247969.1A CN110196152A (en) | 2019-03-29 | 2019-03-29 | The method for diagnosing faults and system of large-scale landscape lamp group based on machine vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110196152A true CN110196152A (en) | 2019-09-03 |
Family
ID=67751760
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910247969.1A Pending CN110196152A (en) | 2019-03-29 | 2019-03-29 | The method for diagnosing faults and system of large-scale landscape lamp group based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110196152A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110749601A (en) * | 2019-10-24 | 2020-02-04 | 中国民用航空总局第二研究所 | Airport runway lamp detection system and method based on images |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007116670A (en) * | 2005-09-20 | 2007-05-10 | Fuji Xerox Co Ltd | Failure diagnostic system and method, and, image forming apparatus |
CN103034838A (en) * | 2012-12-03 | 2013-04-10 | 中国人民解放军63963部队 | Special vehicle instrument type identification and calibration method based on image characteristics |
CN105513038A (en) * | 2014-10-20 | 2016-04-20 | 网易(杭州)网络有限公司 | Image matching method and mobile phone application test platform |
CN108734690A (en) * | 2018-03-02 | 2018-11-02 | 苏州汉特士视觉科技有限公司 | A kind of defects of vision detection device and its detection method |
CN109443709A (en) * | 2018-12-04 | 2019-03-08 | 送飞实业集团有限公司 | It is a kind of to use unmanned plane airfield light light-intensity test system and its control method |
-
2019
- 2019-03-29 CN CN201910247969.1A patent/CN110196152A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007116670A (en) * | 2005-09-20 | 2007-05-10 | Fuji Xerox Co Ltd | Failure diagnostic system and method, and, image forming apparatus |
CN103034838A (en) * | 2012-12-03 | 2013-04-10 | 中国人民解放军63963部队 | Special vehicle instrument type identification and calibration method based on image characteristics |
CN105513038A (en) * | 2014-10-20 | 2016-04-20 | 网易(杭州)网络有限公司 | Image matching method and mobile phone application test platform |
CN108734690A (en) * | 2018-03-02 | 2018-11-02 | 苏州汉特士视觉科技有限公司 | A kind of defects of vision detection device and its detection method |
CN109443709A (en) * | 2018-12-04 | 2019-03-08 | 送飞实业集团有限公司 | It is a kind of to use unmanned plane airfield light light-intensity test system and its control method |
Non-Patent Citations (2)
Title |
---|
何东健等: "《数字图像处理(第三版)》", 28 February 2015 * |
张广渊等: "《数字图像处理基础及OpenCV实现》", 31 December 2014 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110749601A (en) * | 2019-10-24 | 2020-02-04 | 中国民用航空总局第二研究所 | Airport runway lamp detection system and method based on images |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN111259892B (en) | Inspection method, inspection device, inspection equipment and inspection medium for state of indicator lamp | |
KR102229594B1 (en) | Display screen quality detection method, device, electronic device and storage medium | |
US20200380899A1 (en) | Method and apparatus for detecting peripheral circuit of display screen, electronic device, and storage medium | |
CN112395928A (en) | Method for automatically detecting equipment state operation | |
CN107103330A (en) | A kind of LED status recognition methods and device | |
CN112364740B (en) | Unmanned aerial vehicle room monitoring method and system based on computer vision | |
CN117474357B (en) | Power distribution room operation and maintenance management method and system based on deep learning | |
CN107895362A (en) | A kind of machine vision method of miniature binding post quality testing | |
CN112560816A (en) | Equipment indicator lamp identification method and system based on YOLOv4 | |
CN109839386B (en) | Intelligent camera shooting identification system | |
CN109064446A (en) | Display screen quality determining method, device, electronic equipment and storage medium | |
CN205538710U (en) | Inductance quality automatic check out system based on machine vision | |
WO2020007119A1 (en) | Display screen peripheral circuit detection method and device, electronic device and storage medium | |
CN105096305A (en) | Method and device for analyzing state of insulator | |
CN109087281A (en) | Display screen peripheral circuit detection method, device, electronic equipment and storage medium | |
CN117114420B (en) | Image recognition-based industrial and trade safety accident risk management and control system and method | |
CN110618129A (en) | Automatic power grid wire clamp detection and defect identification method and device | |
CN110196152A (en) | The method for diagnosing faults and system of large-scale landscape lamp group based on machine vision | |
CN113850773A (en) | Detection method, device, equipment and computer readable storage medium | |
CN110543809A (en) | Electric power operation site risk supervision intrusion video identification method | |
CN113065462A (en) | Monitoring method, device, equipment and storage medium for power grid overhead line | |
CN109492651B (en) | Intelligent identification method for equipment signal lamp | |
CN116665419A (en) | Intelligent fault early warning system and method based on AI analysis in power production operation | |
CN113034021A (en) | Power transmission line defect analysis method and system based on machine patrol data | |
CN113780224A (en) | Transformer substation unmanned inspection method and system |
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 |