CN107784634A - A kind of power transmission line shaft tower Bird's Nest recognition methods based on template matches - Google Patents
A kind of power transmission line shaft tower Bird's Nest recognition methods based on template matches Download PDFInfo
- Publication number
- CN107784634A CN107784634A CN201710794666.2A CN201710794666A CN107784634A CN 107784634 A CN107784634 A CN 107784634A CN 201710794666 A CN201710794666 A CN 201710794666A CN 107784634 A CN107784634 A CN 107784634A
- Authority
- CN
- China
- Prior art keywords
- transmission line
- image
- template
- shaft tower
- power transmission
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 29
- 244000000626 Daucus carota Species 0.000 title claims abstract description 20
- 230000005540 biological transmission Effects 0.000 title claims abstract description 20
- 235000005770 birds nest Nutrition 0.000 title claims abstract description 20
- 235000005765 wild carrot Nutrition 0.000 title claims abstract description 20
- 238000001514 detection method Methods 0.000 claims abstract description 7
- 238000001914 filtration Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 2
- 230000002401 inhibitory effect Effects 0.000 claims description 2
- 230000000750 progressive effect Effects 0.000 claims description 2
- 238000005516 engineering process Methods 0.000 abstract description 6
- 230000005611 electricity Effects 0.000 abstract description 4
- 238000011160 research Methods 0.000 abstract description 4
- 230000007547 defect Effects 0.000 abstract description 3
- 239000012212 insulator Substances 0.000 abstract description 3
- 239000000725 suspension Substances 0.000 abstract 1
- 230000011218 segmentation Effects 0.000 description 5
- 238000012360 testing method Methods 0.000 description 3
- 239000006002 Pepper Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000012423 maintenance Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000008676 import Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 210000000697 sensory organ Anatomy 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- 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
- G06T2207/10024—Color image
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Computing Systems (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Image Processing (AREA)
Abstract
The defects of research for being mostly based on unmanned plane polling transmission line technology is all the suspension of stranded and foreign matter, the insulator missing on wire is detected.And the research for the horizontal Bird's Nest identification technology of power transmission line shaft tower is few.And picture background is complex, it is difficult to which more satisfied Detection results are made us in acquisition.Shaft tower is the important component in transmission line of electricity, once failure hidden danger, will directly threaten high-voltage fence safe.The present invention relates to a kind of power transmission line shaft tower Bird's Nest recognition methods based on template matches, the figure that will take photo by plane first is converted to HSI spaces from GRB spaces, in order to reduce operand, improves recognition speed, by carrying out dimension-reduction treatment to picture, then pre-processed respectively on H and channel S.After pretreatment, ready template is loaded into, image to be identified is carried out into the information area with template image using template matching method is superimposed, and carries out matched pixel statistics to the image after superposition, so as to obtain matching factor, it is best match to take matching factor maximum.
Description
Technical field
The present invention relates to template matches, and in particular to a kind of power transmission line shaft tower Bird's Nest recognition methods based on template matches.
Background technology
Economic development not only makes urban and rural power grids load rapid growth, also power supply reliability and power supply quality is proposed higher
Requirement.The power circuit corridor in China, it is often necessary to pass through various complicated geographical environments, frequently by lake and reservoir with
And high and steep mountains etc., this coverage of transmission line of electricity is big, distributed areas are wide, transmission range is long, geographical conditions are complicated and changeable and
The features such as notable is influenceed by amblent air temperature, great challenge is brought to the day-to-day operation of circuit, maintenance and maintenance.
The tour of transmission line of electricity typically uses manual patrol mode, though this method is simple, it is less efficient, the cycle compared with
It is long, and need to be equipped with a large amount of optical devices and quality is high, veteran track walker, the requirement to manpower, financial resources is higher.
With the development and application of polling transmission line technology of the China based on unmanned plane, for how under the natural background of complexity,
Using image processing techniques, line facility (such as wire, insulator) is automatically and accurately extracted from aviation image, it is accurate to know
Its defect is not detected, turns into a key technical problem.
At present, the research for being mostly based on unmanned plane polling transmission line technology is hanged on the stranded and foreign matter of wire
The defects of extension, insulator missing, is detected.And the research for the horizontal Bird's Nest identification technology of power transmission line shaft tower is few.Meanwhile
Because picture background is complex, it is difficult to which more satisfied Detection results are made us in acquisition.Shaft tower is the important portion in transmission line of electricity
Part, once failure hidden danger, will directly threaten high-voltage fence safe, or even cause loss difficult to the appraisal.
Herein for this deficiency, the identification of the shaft tower Bird's Nest in primary study Aerial Images.
The content of the invention
The invention discloses a kind of power transmission line shaft tower Bird's Nest recognition methods based on Aerial Images, main contents include taking photo by plane
Image preprocessing, Bird's Nest, display matching result are detected using template matching method.
The present invention overall design philosophy be:Because picture background of taking photo by plane is more complicated, and shooting angle is not fixed, herein
The figure that will take photo by plane first is converted to HSI spaces from GRB spaces.In order to reduce operand, recognition speed is improved, by entering to picture
Row dimension-reduction treatment, then pre-processed respectively on H and channel S.After pretreatment, ready template is loaded into, utilizes mould
Image to be identified is carried out the information area with template image and is superimposed by plate matching method, and matched pixel system is carried out to the image after superposition
Meter, so as to obtain matching factor, it is best match to take matching factor maximum.Fig. 1 is this method overview flow chart.
Brief description of the drawings
Fig. 1 main program flow charts;
Fig. 2-1 plus salt-pepper noise figure after adaptive-filtering with contrasting sectional drawing;
Fig. 2-2 HIS single channel images;
Fig. 2-3 HIS single channel image Threshold segmentations;
Common factor effect is taken after Fig. 2-4 H and channel S Threshold segmentation;
Fig. 2-5 Sobel operators;
Fig. 2-6 Sobel rim detection design sketch;
Fig. 3-1 template matches processes;
Fig. 3-2 Bird's Nest templates;
Fig. 3-3 Bird's Nest recognition results.
Embodiment
Image preprocessing
Due in the absence of perfect state, unavoidably always introducing various noises in image process is obtained, not only hindering
Sense organ, can more hinder the understanding and analysis of subsequent figure source information, and error is caused to result, therefore, to image procossing it
Can gained target image progress denoising as previous.Filtering is the concept in signal transacting, it is therefore an objective to by certain wave in signal
The frequency of section filters out, and is processing method very classical in denoising.
Adaptive median filter
Medium filtering effect depends on the size of filter window, makes very much edge blurry greatly, and too small then denoising effect is bad.
Because noise spot and marginal point are equally the more violent pixels of grey scale change, when common medium filtering changes noise spot gray scale,
Also the gray value of edge pixel will be changed to a certain degree.And noise spot pixel value is nearly all the extreme value in neighborhood, but edge leads to
Chang Buhui is then to limit medium filtering using this feature.
Specifically improved method can be:Progressive scanning picture, when handling each pixel, judge whether the pixel is filter
The maximum or minimum of the lower neighborhood territory pixel of ripple window covering.If it is, using the normal median filter process picture
Element;If it is not, then disregard.This method can effectively remove burst noise point, especially salt-pepper noise,
And have little influence on edge.Comparison diagram is as shown in Fig. 2-1.
In the present invention, because its adjacent spots has very strong correlation, edge feature ensures not to be blurred again, so in
Value filtering method is most suitable.
Spatial alternation and processing
HSI color spaces describe color with tone (Hue), color saturation (Saturation) and brightness (Intensity)
It is color.The classification and depth degree of color are represented with tone and saturation degree, the relative shading value of color is indicated with brightness.HSI colors
Space separates the colourity of image and brightness, is provided a great convenience for Color Image Processing, for particular color, only needs
H and S components are directed to, are analyzed and processed in plane, and ignore the I component on right side.Aerial Images are transformed into HSI spaces,
And separate triple channel, H and S component single channel images are obtained as shown in Fig. 2-2.
Because image resolution ratio is higher, the time of consuming is calculated with regard to long.Therefore, after channel separation, to image
Size is reset, and is reduced to original 0.5 times, can so reduce amount of calculation, accelerates code operational efficiency.Then respectively to H
Enter row threshold division with channel S, the foreground image of acquisition is as Figure 2-3.
From Fig. 2-3, H passage noises are relatively fewer, also essentially eliminate light intensity in single image and prospect is carried
The influence taken, and substantial amounts of noise color lump in channel S be present.Pass through a large amount of tests to different colours shaft tower in different background
Understand, H component single channel Threshold segmentations, shaft tower in most cases can more fully be extracted;Though S components can carry shaft tower
Take, flase drop but easily occur.It can be seen that a certain component of HSI color spaces is used alone, it is difficult to obtain accurate prospect
Image.Two images shown in Fig. 2-3 are sought common ground, can so utilize H and S component informations simultaneously, while ignore I component,
The influence of illumination is avoided, as in Figure 2-4.
More than analysis understand, image is transformed into HSI spaces from rgb color space first, then H and channel S are entered respectively
Row threshold division, finally segmentation result is sought common ground, this method can be reduced meadow, the first-class background removal of concrete floor dry
Disturb the factor of identification.
Sobel rim detections
Sobel operators are a kind of edge detection methods based on gradient.The expression formula of Sobel operators is:
G (i, j)=| f (i-l, j+1)+2f (i, j+l)+f (i+1, j+1)-f (i-1, j-l) -2f (i, j-l)-f (i+1,
j-l)|
+|f(i-l,j-1)+2f(i-l,j)+f(i-l,j+l)-f(i+1,j-l)-2f(i+1,j)-f(i+l,j+
1)|
Two convolutional calculation templates of Sobel edge detection operators as shown in Figure 2-5, each point in image with this two
Individual template makees convolution, and first template responds maximum to common vertical edge, and second template responds most to horizontal edge
Greatly.Output valve of the maximum of two convolution as the point, operation result are a breadths edge magnitude images.Sobel operators are to ash
Degree gradual change and the more image procossing of noise obtain preferably.
Sobel operators are smoothed to image first, have certain noise inhibiting ability, then remake differential fortune
Calculate, good edge effect can be produced, but also will detect that some pseudo-edges so that edge is thicker.It is real-time in view of image
Processing requirement calculating speed is fast, and the feature of shaft tower Bird's Nest is obvious, and required precision is not high, therefore using based on first derivative
Sobel operators carry out rim detection.Testing result is as shown in figures 2-6.
Bird's Nest identification based on template matches
Matching is to be combined the expression inputted with oneself has in advance, or the mistake of corresponding pattern is found according to known mode
Journey, that is, establish it is unknown with it is known between contact to identify the process of unknown object.The collection of known mode is collectively referred to as template
Storehouse, the set of unknown pattern turn into test sample storehouse.As shown in figure 3-1, the class belonging to pattern to be sorted is found in ATL
Other process is to match.
Template imports
The present invention is identified for invoice number, and the digital template of the invoice number font of standard is imported into database
In, image input language, i.e. sentence are carried using Matlab, i.e.,:
Temp=imread (' C:\Users\Administrator\Desktop\muban.bmp');
Bird's Nest template is as shown in figure 3-2.
Bird's Nest identification process
First by font masters input database to be identified, then picture pic to be identified of taking photo by plane is inputted in program,
By spatial alternation, Threshold segmentation, rim detection, image background is removed, and is multiplied with template, counts each template
Coefficient correlation after matching, coefficient correlation is bigger to represent that matching degree is higher.The maximum point of coefficient correlation is chosen to tie as identification
Fruit, finally recognition result is exported.Recognition result is as shown in Fig. 3-3.
The specific embodiment of the present invention is described above.It is to be appreciated that the invention is not limited in above-mentioned
Particular implementation, those skilled in the art can make various deformations or amendments within the scope of the claims, this not shadow
Ring the substantive content of the present invention.
Claims (5)
- A kind of 1. power transmission line shaft tower Bird's Nest recognition methods based on template matches, it is characterised in that:The figure that will take photo by plane first is empty from GRB Between be converted to HSI spaces;Secondly, by carrying out dimension-reduction treatment to picture, then pre-processed respectively on H and channel S;By After pretreatment, ready template is loaded into, image to be identified is carried out into the information area with template image using template matching method folds Add, matched pixel statistics is carried out to the image after superposition, so as to obtain matching factor, it is optimal to take matching factor maximum Match somebody with somebody.
- 2. power transmission line shaft tower Bird's Nest recognition methods according to claim 1, it is characterised in that:Described being filtered into is adaptive Medium filtering, progressive scanning picture, when handling each pixel, judge whether the pixel is the lower neighborhood picture of filter window covering The maximum or minimum of element;If it is, using the normal median filter process pixel;If it is not, then not locate Reason.
- 3. power transmission line shaft tower Bird's Nest recognition methods according to claim 1, it is characterised in that:Described spatial alternation is: HSI color spaces separate the colourity of image and brightness, are provided a great convenience for Color Image Processing, for specific face Color, it is only necessary to for H and S components, analyzed and processed in plane, and ignore the I component on right side.
- 4. power transmission line shaft tower Bird's Nest recognition methods according to claim 1, it is characterised in that:Sobel rim detections are specific For:Sobel operators are smoothed to image first, are had certain noise inhibiting ability, are then remake and differentiate, can To produce good edge effect.
- 5. power transmission line shaft tower Bird's Nest recognition methods according to claim 1, it is characterised in that:Described template matches are specific For:By in font masters input database to be identified, then picture to be identified of taking photo by plane is inputted in program, by a series of pre- places Reason operation removes image background, and is multiplied with template, counts the coefficient correlation after each template matches, coefficient correlation is got over It is big to represent that matching degree is higher.The maximum point of coefficient correlation is chosen as recognition result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710794666.2A CN107784634A (en) | 2017-09-06 | 2017-09-06 | A kind of power transmission line shaft tower Bird's Nest recognition methods based on template matches |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710794666.2A CN107784634A (en) | 2017-09-06 | 2017-09-06 | A kind of power transmission line shaft tower Bird's Nest recognition methods based on template matches |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107784634A true CN107784634A (en) | 2018-03-09 |
Family
ID=61437957
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710794666.2A Pending CN107784634A (en) | 2017-09-06 | 2017-09-06 | A kind of power transmission line shaft tower Bird's Nest recognition methods based on template matches |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107784634A (en) |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108759670A (en) * | 2018-05-31 | 2018-11-06 | 成都唐源电气股份有限公司 | A kind of contact line abrasion device for dynamically detecting based on non-contact detection technology |
CN108764020A (en) * | 2018-03-30 | 2018-11-06 | 广东工业大学 | A kind of Bird's Nest recognition methods on high tension electric tower based on unmanned plane image |
CN109445432A (en) * | 2018-10-31 | 2019-03-08 | 中国科学技术大学 | Unmanned plane and ground mobile robot formation localization method based on image |
CN109544501A (en) * | 2018-03-22 | 2019-03-29 | 广东电网有限责任公司清远供电局 | A kind of transmission facility defect inspection method based on unmanned plane multi-source image characteristic matching |
CN110020598A (en) * | 2019-02-28 | 2019-07-16 | 中电海康集团有限公司 | A kind of method and device based on foreign matter on deep learning detection electric pole |
CN110276241A (en) * | 2019-03-28 | 2019-09-24 | 广东工业大学 | A kind of stockbridge damper recognition methods based on template matching |
CN110516531A (en) * | 2019-07-11 | 2019-11-29 | 广东工业大学 | A kind of recognition methods of the dangerous mark based on template matching |
CN113050153A (en) * | 2020-11-06 | 2021-06-29 | 泰州物族信息科技有限公司 | Object presence validity identification platform |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105536205A (en) * | 2015-12-08 | 2016-05-04 | 天津大学 | Upper limb training system based on monocular video human body action sensing |
CN105562361A (en) * | 2015-12-23 | 2016-05-11 | 西安工程大学 | Independent sorting method of fabric sorting robot |
KR101684139B1 (en) * | 2015-06-30 | 2016-12-08 | 주식회사 버츄얼스톰 | Augmented reality implementation method using multiple block layer light enviroment |
CN106326808A (en) * | 2015-06-23 | 2017-01-11 | 上海深邃智能科技有限公司 | Method for detecting bird nests in power transmission line poles based on unmanned plane images |
CN106683075A (en) * | 2016-11-22 | 2017-05-17 | 广东工业大学 | Power transmission line tower cross arm bolt defect detection method |
CN106980827A (en) * | 2017-03-16 | 2017-07-25 | 天津大学 | A kind of method of Bird's Nest in identification transmission line of electricity based on Aerial Images |
-
2017
- 2017-09-06 CN CN201710794666.2A patent/CN107784634A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106326808A (en) * | 2015-06-23 | 2017-01-11 | 上海深邃智能科技有限公司 | Method for detecting bird nests in power transmission line poles based on unmanned plane images |
KR101684139B1 (en) * | 2015-06-30 | 2016-12-08 | 주식회사 버츄얼스톰 | Augmented reality implementation method using multiple block layer light enviroment |
CN105536205A (en) * | 2015-12-08 | 2016-05-04 | 天津大学 | Upper limb training system based on monocular video human body action sensing |
CN105562361A (en) * | 2015-12-23 | 2016-05-11 | 西安工程大学 | Independent sorting method of fabric sorting robot |
CN106683075A (en) * | 2016-11-22 | 2017-05-17 | 广东工业大学 | Power transmission line tower cross arm bolt defect detection method |
CN106980827A (en) * | 2017-03-16 | 2017-07-25 | 天津大学 | A kind of method of Bird's Nest in identification transmission line of electricity based on Aerial Images |
Non-Patent Citations (2)
Title |
---|
丁巧: "机器视觉激光焊接缺陷检测算法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 * |
张少平: "输电线路典型目标图像识别技术研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109544501A (en) * | 2018-03-22 | 2019-03-29 | 广东电网有限责任公司清远供电局 | A kind of transmission facility defect inspection method based on unmanned plane multi-source image characteristic matching |
CN108764020A (en) * | 2018-03-30 | 2018-11-06 | 广东工业大学 | A kind of Bird's Nest recognition methods on high tension electric tower based on unmanned plane image |
CN108759670A (en) * | 2018-05-31 | 2018-11-06 | 成都唐源电气股份有限公司 | A kind of contact line abrasion device for dynamically detecting based on non-contact detection technology |
CN109445432A (en) * | 2018-10-31 | 2019-03-08 | 中国科学技术大学 | Unmanned plane and ground mobile robot formation localization method based on image |
CN110020598A (en) * | 2019-02-28 | 2019-07-16 | 中电海康集团有限公司 | A kind of method and device based on foreign matter on deep learning detection electric pole |
CN110276241A (en) * | 2019-03-28 | 2019-09-24 | 广东工业大学 | A kind of stockbridge damper recognition methods based on template matching |
CN110516531A (en) * | 2019-07-11 | 2019-11-29 | 广东工业大学 | A kind of recognition methods of the dangerous mark based on template matching |
CN110516531B (en) * | 2019-07-11 | 2023-04-11 | 广东工业大学 | Identification method of dangerous goods mark based on template matching |
CN113050153A (en) * | 2020-11-06 | 2021-06-29 | 泰州物族信息科技有限公司 | Object presence validity identification platform |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107784634A (en) | A kind of power transmission line shaft tower Bird's Nest recognition methods based on template matches | |
CN109800631A (en) | Fluorescence-encoded micro-beads image detecting method based on masked areas convolutional neural networks | |
CN104715239B (en) | A kind of vehicle color identification method based on defogging processing and weight piecemeal | |
CN108133481A (en) | A kind of image processing algorithm for fluorescence immune chromatography strip imaging system | |
CN106683075A (en) | Power transmission line tower cross arm bolt defect detection method | |
CN113989662B (en) | Remote sensing image fine-grained target identification method based on self-supervision mechanism | |
CN110211101A (en) | A kind of rail surface defect rapid detection system and method | |
CN108550159B (en) | Flue gas concentration identification method based on image three-color segmentation | |
CN103235830A (en) | Unmanned aerial vehicle (UAV)-based electric power line patrol method and device and UAV | |
CN107492094A (en) | A kind of unmanned plane visible detection method of high voltage line insulator | |
CN109685766A (en) | A kind of Fabric Defect detection method based on region fusion feature | |
Chao et al. | High throughput spectral imaging system for wholesomeness inspection of chicken | |
CN107818303A (en) | Unmanned plane oil-gas pipeline image automatic comparative analysis method, system and software memory | |
CN103914707B (en) | Green channel product auxiliary discriminating method based on support vector machine | |
CN109613023A (en) | A kind of fruit surface defect rapid detection method of regional luminance adaptively correcting | |
CN105974120A (en) | Automatic detection device and method for C-reactive protein chromaticity | |
CN107704878A (en) | A kind of high-spectral data storehouse semi-automation method for building up based on deep learning | |
CN111007039A (en) | Automatic extraction method and system for sub-pixel level water body of medium-low resolution remote sensing image | |
CN107462182A (en) | A kind of cross section profile deformation detecting method based on machine vision and red line laser | |
CN106251328A (en) | A kind of method utilizing two dimension high spectrum image identification pollutants on surface of whole chicken rinses | |
CN110298410A (en) | Weak target detection method and device in soft image based on deep learning | |
CN110276241A (en) | A kind of stockbridge damper recognition methods based on template matching | |
CN113158969A (en) | Apple appearance defect identification system and method | |
CN105139034B (en) | A kind of ship detecting method that combination spectrum filters out | |
CN106875407A (en) | A kind of unmanned plane image crown canopy dividing method of combining form and marking of control |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180309 |