CN204718967U - The accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process - Google Patents
The accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process Download PDFInfo
- Publication number
- CN204718967U CN204718967U CN201520401373.XU CN201520401373U CN204718967U CN 204718967 U CN204718967 U CN 204718967U CN 201520401373 U CN201520401373 U CN 201520401373U CN 204718967 U CN204718967 U CN 204718967U
- Authority
- CN
- China
- Prior art keywords
- paper tube
- chemical fiber
- image
- fiber paper
- trigger
- 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.)
- Withdrawn - After Issue
Links
Abstract
The utility model discloses the accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process, this system comprises an auxiliary sign thing, is arranged at detected chemical fiber paper tube end face; One detects video camera, carries out cylindrical picture signals collecting, a PC to detected chemical fiber paper tube, is inside equipped with picture signal training and repairs software, and this PC provides the collection control signal after training is repaired to detection video camera; One trigger, trigger rotates according to paper tube and exports trigger pip in real time, and the detected paper tube cylinder of trigger pip correspondence is just to the region detecting video camera, and the rising edge of the trigger pip that trigger sends corresponds to the moment that detected paper tube cylindrical surface areas is rotated in place.The problems such as it can solve geometric distortion that traditional image exists, degree of accuracy is not high, image mosaic effect is undesirable; Solve preferably again and this difficult problem is accurately spliced to zones of different post panoramic picture, have that image procossing precision is high, calculated amount is little, antijamming capability is strong and be easy to the features such as realization.
Description
Technical field
The utility model relates to the intelligent image treatment technology used in the detection technique of industrial products, more specifically refers to the accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process adopted in detecting for chemical fiber paper tube outward appearance.
Background technology
Chemical fiber paper tube is mainly used in textile industry, for high-speed winding or add play spindle basic guarantee is provided.Whether its external form physical dimension, circularity, concentricity and appearance have flaw to be paper tube very important parameter when serving as intermediate medium, and current paper tube manufacturing enterprise generally adopts the means of manual detection, ensures that the paper tube outward appearance of output conforms to quality requirements.But there is following drawback in manual detection method:
1, inconsistent due to subjective judgement standard between testing staff, easily causes the inconsistent of testing result;
2, the product easily caused due to the physiological fatigue of testing staff self is undetected, flase drop;
The factors such as 3, human cost rises year by year, and mobility of people is large, cause manual detection cost more and more higher.
The applicant, in order to solve the problem, proposes a kind of chemical fiber paper tube appearance delection device and method, and the proposition Chinese patent application on April 8th, 2015 with regard to correlation technique, application number is 201510161827.5.This application solves paper tube outward appearance preferably and accurately screens, and detects paper tube of different nature, and classifies in real time.Be applicable to paper tube apparent parameter testing process, applied for machines vision technique and automatic control technology realize the object of high precisely detection, criteria classification screening, increase work efficiency, reduce the cost of labour.
But, in the detection application of reality, how to detect after image process, and the geometric distortion how overcoming traditional image existence is more responsive, the problem that easily causes that splicing result is inaccurate, degree of accuracy is not high, image mosaic effect is undesirable etc., be in the application institute in the face of and problem to be solved.
The splicing of image mainly realizes according to the similarity of two width image overlapping regions, generally there is the method based on region and feature based at present, traditional method is all generally the splicing that directly corresponding from image overlapping region grey scale pixel value or chromatic value carry out considering image, the shortcoming that during this method ubiquity splicing, calculated amount is large, and the geometric distortion that result of calculation causes the change of illumination and camera lens is more responsive, therefore easily cause the problems such as splicing result is inaccurate, degree of accuracy is not high, image mosaic effect is undesirable.
Need the zones of different cylindrical picture collected accurately to splice in based on the chemical fiber paper tube outward appearance detection system of machine vision, to eliminate the region of repeated acquisition, follow-up Digital Image Processing and analysis, accurately to splice this difficult problem to obtained panoramic picture, be pendulum in the detection industry manufacturer of industrial products in intelligent image treatment technology process, institute expects the problem of solution.
Utility model content
For the problems referred to above, task of the present utility model is to provide the accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process, to solve in traditional image procossing, the ubiquity shortcoming that calculated amount is large when splicing, and the geometric distortion that result of calculation causes the change of illumination and camera lens is more responsive, easily cause that splicing result is inaccurate, degree of accuracy is not high, image mosaic effect is the problem such as undesirable.
In order to complete above-mentioned task, the utility model adopts following technical scheme:
The accurate splicing system of image in this chemical fiber paper tube outward appearance testing process,
The accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process, this system comprises:
One auxiliary sign thing, is arranged at detected chemical fiber paper tube end face;
One detects video camera, and this video camera carries out cylindrical picture signals collecting to detected chemical fiber paper tube, and the picture signal after gathering is outputted to PC;
One PC, is inside equipped with picture signal training and repairs software, and this PC provides the collection control signal after training is repaired to detection video camera;
One trigger, trigger rotates according to paper tube and exports trigger pip in real time, the detected paper tube cylinder of each trigger pip correspondence is just to the region detecting video camera, and the rising edge of the trigger pip that trigger sends corresponds to the moment that detected paper tube cylindrical surface areas is rotated in place.
Described auxiliary sign thing has relevant diameter to detected chemical fiber paper tube and coaxially arranges, and the auxiliary sign that this auxiliary sign thing is arranged is black and white grid and corresponding by even partition part number with detected chemical fiber paper tube end face circumference.
Described PC generates according to the trigger pip of trigger and exports and gathers control signal, controls the picture signal of the chemical fiber paper tube cylinder corresponding region that video camera Real-time Collection is detected.
In chemical fiber paper tube outward appearance testing process of the present utility model in the accurate disposal system of image, owing to have employed above-mentioned technical scheme, by an auxiliary sign thing, be arranged at detected chemical fiber paper tube end face; One detects video camera, and this video camera carries out cylindrical picture signals collecting to detected chemical fiber paper tube, and the picture signal after gathering is outputted to PC; One PC, is inside equipped with picture signal training and repairs software, and this PC provides the collection control signal after training is repaired to detection video camera; One trigger, trigger rotates according to paper tube and exports trigger pip in real time, the detected paper tube cylinder of each trigger pip correspondence is just to the region detecting video camera, and the rising edge of the trigger pip that trigger sends corresponds to the moment that detected paper tube cylindrical surface areas is rotated in place.The problems such as the utility model can solve geometric distortion that traditional image exists, degree of accuracy is not high, image mosaic effect is undesirable; Solve preferably again and this difficult problem is accurately spliced to zones of different post panoramic picture, have that image procossing precision is high, calculated amount is little, antijamming capability is strong and be easy to the features such as realization.
Accompanying drawing explanation
Fig. 1 is the auxiliary sign schematic diagram that the utility model is arranged at auxiliary sign thing.
Fig. 2 is the position relationship schematic diagram of auxiliary sign thing of the present utility model and detected chemical fiber paper tube.
Fig. 3 is auxiliary sign border of the present utility model schematic diagram.
Fig. 4 is the position of center line schematic diagram of auxiliary sign thing of the present utility model and detected chemical fiber paper tube.
Fig. 5 is in Fig. 4, the schematic diagram of center line on some auxiliary signs.
Fig. 6 is in Fig. 4, the situation of center line between two auxiliary signs, and center line is carried out the schematic diagram of translation.
Fig. 7 is the accurate disposal system schematic block diagram of image in the utility model chemical fiber paper tube outward appearance testing process.
Fig. 8 is the utility model trigger pip waveform schematic diagram.
Fig. 9 is the principle schematic that the utility model trigger pip gathers paper tube cylinder.
Figure 10 is in the utility model, waveform schematic diagram during the loss of trigger pip.
Figure 11 is in the utility model, waveform schematic diagram during the interference of trigger pip.
Figure 12 is in the utility model, the signal window waveform schematic diagram before and after training.
Encoded radio schematic diagram on the center line that Figure 13 is detected chemical fiber paper tube adjacent image corresponding to auxiliary sign.
Figure 14 is the beginning boundary and the end boundary encoded radio schematic diagram that calculate adjacent image.
Figure 15 is image mosaic schematic diagram after employing the utility model system.
Figure 16 is in the utility model embodiment, the trigger pip trigger point distribution schematic diagram of detected chemical fiber paper tube cylinder setting.
Figure 17 is in the utility model embodiment, auxiliary sign schematic diagram.
Figure 18 is in the utility model embodiment, identifies the encoded radio schematic diagram on adjacent image center line.
Figure 19 is in the utility model embodiment, the encoded radio schematic diagram of beginning boundary and end boundary during splicing.
Figure 20 is in the utility model embodiment, the schematic diagram after image mosaic.
Embodiment
As shown in Figure 1, in the utility model system, the auxiliary sign on auxiliary sign thing is made up of m grid of vertical direction, wherein each quadrille " black " and " in vain " two kinds of method for expressing, and corresponding binary value is " 1 " and " 0 ".The binary value that auxiliary sign is corresponding is called the encoded radio of this auxiliary sign.
As shown in Figure 2, the position of its cylinder zones of different can be determined.If paper tube 20 end face diameter is D, girth is L, has L=D* π; If the width of the auxiliary sign of auxiliary sign thing 50 is w, n is the number of auxiliary sign, make w*n=L=D* π.N auxiliary sign is corresponding with paper tube 20 end face circumference, namely end face circumference is evenly divided into n part, the corresponding portion wherein of each auxiliary sign, also the relevant position of paper tube cylinder is corresponded to, achieve the location of cylinder, n is larger, and the width w=D π/n of corresponding each auxiliary sign is less, and positioning precision is higher; Otherwise positioning precision is lower.But the width w of auxiliary sign is too little is unfavorable for its gray-scale value of image procossing identification on the contrary, easily occurs identification error.Therefore positioning precision and image recognition sensitivity is generally taken into account in practical application to determine w value.The height h of each grid of auxiliary sign artificially determines, generally makes h=w, makes each grid of auxiliary sign be a square, is convenient to carry out image recognition.
Corresponding n the auxiliary sign of paper tube cylinder, each auxiliary sign is made up of m grid, and m and n should meet relational expression 2
m>=n.Wherein 2
mduring=n, represent 2
mkind of auxiliary sign is just corresponding n position of end face circumference.As given n, should choose and make 2
mthat minimum m value that>=n sets up.Such as, if paper tube cylinder is evenly divided into 1000 parts (n=1000), then as m=10 2
10=1024 slightly larger than 1000, then the auxiliary sign of this paper tube forms most suitable by 10 grid (m=10).
As shown in Figure 3, for the ease of distinguishing auxiliary sign coding, one group of auxiliary sign beginning flag grid is set in the starting position and end position that identify auxiliaring coding, h is identical with the feature of auxiliary sign for beginning flag grid wide w height, also there are binary two kinds of method for expressing, i.e. " black " representative " 1 " and " in vain " representative " 0 ".Corresponding n the auxiliary sign of paper tube cylinder also has n beginning flag grid, and the quantity of a beginning flag grid does not limit, but generally with black white two beginning flag grid for the best.
Fault-tolerant and the error correcting capability that auxiliary sign possesses when locating.The pattern of auxiliary sign easily causes identification error due to reasons such as the deformation of light-illuminating change, image, surface are stained, realizes the fault-tolerant of encoded radio and error correction in the application by the encoding characteristics between auxiliary sign self-contrast degree and contiguous tokens.
Several auxiliary signs are all included for the every piece image collected, makes the encoded radio linearly aligned transfer of its adjacent auxiliary sign.First, the auxiliary sign on recognition image center line; Then the auxiliary sign encoded radio of optional position on image can be extrapolated according to the linear array relation of adjacent auxiliary sign.The center line 60 of image refers to the set to two adjacent side distance identical pointses on image, as shown in Figure 4.
As shown in Figure 5, be that benchmark chooses image-region highly one by one by w width and h with center line, identify its gray-scale value, this region of expression being greater than 0.5 is mainly black, and its encoded radio is designated as " 1 "; This region of expression being less than 0.5 is mainly white, and its encoded radio is " 0 ".As gray-scale value equals 0.5, illustrate that center line is between two auxiliary signs and adjacent checkered color is different, as shown in Figure 6.This kind of situation, then by center line to left 0.5w, again choose image-region, and identify its gray-scale value.
Gray-scale value as described above by each grid of auxiliary sign self can allow that certain area color is made mistakes, and (width w* height is h) relevant for the area of its fault-tolerant ability and each grid.When the color of auxiliary sign makes mistakes the fault-tolerant ability exceeding himself, by the linear relationship between contiguous tokens error correction carried out to it thus recover correct encoded radio, the steps include:
1. get one group of adjacent n encoded radio and be set to set A, element encoded radio is wherein designated as A
i(i=1,2 ..., n).
2. by the elements A in set A
i(i ≠ 1) and first encoded radio A
1subtract each other, obtain difference set.Element linear for difference is taken out composition set B, element is wherein designated as B
i, i=1,2 ..., k, k≤n-1.If the element number inside set B meets
time, then the encoded radio of set B is correct; If k does not meet above-mentioned condition, then by the A in set A
i(
i≠ 1,2) with second encoded radio A
2subtract each other, obtain difference set.Element linear for difference is taken out composition set C, element is wherein designated as C
i, i=1,2 ..., k, k≤n-1.If the element number inside set C meets
time, then the encoded radio gathering C is correct, if k does not meet above-mentioned condition, then by elements A in set A
i(i ≠ 1,2,3) and the 3rd A
3and judge whether k satisfies condition as mentioned above, until when judging that k satisfies condition, thus find out correct encoded radio.
3. there is the set of correct encoded radio for benchmark with above-mentioned, other encoded radio in set A can be recovered according to the linear relationship between adjacent encoder value.
As shown in Figure 7, the splicing system of the utility model panorama sketch comprises:
Detect video camera 10, this video camera 10 carries out cylindrical picture signals collecting to detected chemical fiber paper tube 20, and the picture signal after gathering is outputted to PC 30;
PC 30, be inside equipped with picture signal training and repair software, this PC 30 carries out training reparation to the picture signal that detected video camera exports;
Trigger 40, trigger 40 rotates according to paper tube 20 and exports trigger pip in real time, detected paper tube 20 cylinder of each trigger pip correspondence is just to the region detecting video camera 10, and the rising edge of the trigger pip that trigger 40 sends corresponds to the moment that detected paper tube 20 cylindrical surface areas is rotated in place.
Trigger 40 rotates according to paper tube 20 and exports trigger pip in real time, and waveform as shown in Figure 8.Each trigger pip is corresponding paper tube cylinder is just to a region of video camera, and the rising edge of trigger pip corresponds to the moment that this cylindrical surface areas is rotated in place, as shown in Figure 9.PC carries out reparation training to the trigger pip of trigger 40, and generates collection control signal, controls the image of video camera 10 Real-time Collection paper tube cylinder corresponding region.In order to ensure that image correctly splices, require all image informations collected should comprise whole cylindrical surface areas and position, cylindrical surface areas corresponding to every width image is accurate.But, because the reason such as mechanical vibration, electromagnetic interference (EMI) causes trigger pip abnormal, the mistake collection of cylindrical picture can be caused.Abnormal trigger pip comprises signal disturbing, signal deletion.As shown in Figure 10, should occur the rising edge of trigger pip, but do not occur in the t2 moment, this kind of situation is called that trigger pip is lost.As shown in figure 11, occurred a trigger pip in t1 and the t2 moment, this kind of situation is called as the interference of trigger pip.
The utility model, for above-mentioned abnormal signal, proposes and automatically trains reparation, and its process is as follows:
1. copied by trigger pip and obtain initial collection control signal;
2. the rising edge time of repeatedly trigger pip and the deration of signal are carried out arithmetic mean, as the collection control signal after training, also referred to as " memory signal ";
3. the undesired signal occurred in trigger pip is filtered by " signal window "; To the loss of trigger pip, memory signal is copied as collection control signal and export.
" signal window " as shown in figure 12, refers to moment the latest that trigger pip corresponding to a certain cylindrical surface areas arrive, mistiming the earliest between the moment.T in figure
fpoint represents the earliest time point that certain trigger pip arrives; t
wpoint represents the time point the latest that this trigger pip arrives; t
zpoint represents the intermediate value of trigger pip the earliest and the latest time point, then
The mistiming in moment and moment is the earliest the width of " signal window " the latest, for | t
w-t
f| individual chronomere.If trigger pip occurs in " signal window ", then copy this trigger pip and obtain collection signal and export; If trigger pip does not occur in " signal window ", represent and occurred dropout or signal disturbing, then export memory signal in the tw moment, to reach the object of repairing collection signal.
" signal window " can adjust according to the change of trigger pip.If the moment appeared at corresponding to the trigger pip in " signal window " is ti at every turn, to n ti calculating mean value t of this trigger pip
average, then
△ t=|t
average-t
z| ... formula (3)
If T is the time threshold of trigger pip change.As △ t<T, illustrate that n trigger pip is all at intermediate value t
zneighbouring appearance, without the need to adjustment " signal window "; As △ t>=T, the average t of n trigger pip is described
averagedepart from t
zlarger.As t
average>t
z, represent that trigger pip is at t
zright side occurs comparatively frequent, otherwise then represents at t
zleft side occurs comparatively frequent, needs " signal window " to do corresponding movement, and time shaft t is formed new " signal window " again, and moment, the latest moment and intermediate value all there occurs change the earliest, i.e. moment t ' the earliest
fmoment t ' the latest
wand intermediate value t '
z, as shown in figure 12.
The utility model is at auxiliary sign and gather on the reparation basis of control signal, further provide a kind of cylindrical picture Panoramagram montage method, its method is described below: first accurately located paper tube cylinder by auxiliary sign, then trains collection control signal and gets rid of dropout and interference.On this basis, whole paper tube cylindrical picture can be gathered accurately, and the auxiliary sign ensureing corresponding to cylinder different images region collects and its encoded radio all can be correctly validated.The encoded radio of auxiliary sign on each width picture centre line is designated as p
i.As shown in figure 13, p
i-1, p
i, p
i+1it is the encoded radio of the auxiliary sign on the i-th-1, i-th and i-th+1 width picture centre line.According to the p of cylindrical picture acquisition order successively recognition image
i, then calculate adjacent image p
idifference.As shown in figure 14, when accurately splicing kth width and k-1 width image, according to p
kand p
k-1the beginning boundary of splicing regions can be calculated
with terminated boundary
As shown in figure 15, adjacent image carries out image mosaic according to starting and ending border, and splicing regions is as shown in fringe area oblique in figure.
To gather 16 paper tube cylindrical pictures, be described as follows:
Paper tube is set and gathers number;
If gather 16 paper tube cylindrical pictures, then 16 trigger pip points are set at equal intervals around paper tube cylinder, as shown in figure 16.
Setting auxiliary sign:
Setting binary coding is auxiliary sign coding (hereinafter referred to as coding), as shown in figure 17 white grid representative " 0 " black lattice representative " 1 ".If paper tube cylindrical picture is divided into n=1000 part, then w=П D/n; 2
m=1000, m=10; During and if only if w=h=П D/n, auxiliary sign grid is just square.
Setting benchmark collection signal, copies by the trigger pip appeared in " signal window " scope the image that the collection signal obtained carries out gathering paper tube cylinder corresponding region.The trigger pip in signal window is not appeared at, with the acquired original signal of computing machine memory for benchmark carries out collection paper tube figure cylinder picture at the appointed time;
Image mosaic, determines the center line capturing image,
Find out and often open binary coding corresponding on picture centre line, as shown in figure 18; Judge that whether the coding on center line is correct, if correctly carry out next step; Carry out if incorrect encoding error detection and finding out correct coding corresponding with it;
The splicing calculating contiguous concatenation image according to formula (3) and formula (4) plays the coding of initial line and end boundary, calculates the start code value of adjacent image splicing regions as shown in figure 19 and terminate encoded radio according to the coding on center line;
As shown in figure 20, accurately the border of splicing adjacent image completes image mosaic.
In sum, in chemical fiber paper tube outward appearance testing process of the present utility model, the accurate disposal system of image is not only applicable to the image procossing that chemical fiber paper tube outward appearance detects, and the panorama sketch being applicable to other similar cylinder products accurately processes.
Those of ordinary skill in the art will be appreciated that, above embodiment is only used to the utility model is described, and be not used as restriction of the present utility model, as long as in spirit of the present utility model, all will drop in Claims scope of the present utility model the change of the above embodiment, modification.
Claims (3)
1. the accurate disposal system of image in chemical fiber paper tube outward appearance testing process, it is characterized in that, this system comprises:
One auxiliary sign thing, is arranged at detected chemical fiber paper tube end face;
One detects video camera, and this video camera carries out cylindrical picture signals collecting to detected chemical fiber paper tube, and the picture signal after gathering is outputted to PC;
One PC, is inside equipped with picture signal training and repairs software, and this PC provides the collection control signal after training is repaired to detection video camera;
One trigger, trigger rotates according to paper tube and exports trigger pip in real time, the detected paper tube cylinder of each trigger pip correspondence is just to the region detecting video camera, and the rising edge of the trigger pip that trigger sends corresponds to the moment that detected paper tube cylindrical surface areas is rotated in place.
2. the accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process as claimed in claim 1, is characterized in that:
Described auxiliary sign thing has relevant diameter to detected chemical fiber paper tube and coaxially arranges, and the auxiliary sign that this auxiliary sign thing is arranged is black and white grid and corresponding by even partition part number with detected chemical fiber paper tube end face circumference.
3. the accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process as claimed in claim 1, is characterized in that:
Described PC generates according to the trigger pip of trigger and exports and gathers control signal, controls the picture signal of the chemical fiber paper tube cylinder corresponding region that video camera Real-time Collection is detected.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201520401373.XU CN204718967U (en) | 2015-06-11 | 2015-06-11 | The accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201520401373.XU CN204718967U (en) | 2015-06-11 | 2015-06-11 | The accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN204718967U true CN204718967U (en) | 2015-10-21 |
Family
ID=54317956
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201520401373.XU Withdrawn - After Issue CN204718967U (en) | 2015-06-11 | 2015-06-11 | The accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN204718967U (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104897688A (en) * | 2015-06-11 | 2015-09-09 | 龚厚志 | Image accurate processing method and system for chemical fiber paper tube appearance detection |
| CN106733680A (en) * | 2016-11-21 | 2017-05-31 | 上海常良智能科技有限公司 | Chemical fiber paper tube classification Transmission system based on image recognition |
| CN107328781A (en) * | 2017-05-23 | 2017-11-07 | 镇江苏仪德科技有限公司 | A kind of columnar product detection method of surface flaw and device based on machine vision |
| CN109454006A (en) * | 2018-10-08 | 2019-03-12 | 杭州慧知连科技有限公司 | Chemical fiber wire ingot is stumbled the device and its detection stage division of a defect on-line checking and classification |
-
2015
- 2015-06-11 CN CN201520401373.XU patent/CN204718967U/en not_active Withdrawn - After Issue
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104897688A (en) * | 2015-06-11 | 2015-09-09 | 龚厚志 | Image accurate processing method and system for chemical fiber paper tube appearance detection |
| CN104897688B (en) * | 2015-06-11 | 2017-08-25 | 龚厚志 | The accurate processing method of image and system in a kind of chemical fiber paper tube outward appearance detection process |
| CN106733680A (en) * | 2016-11-21 | 2017-05-31 | 上海常良智能科技有限公司 | Chemical fiber paper tube classification Transmission system based on image recognition |
| CN106733680B (en) * | 2016-11-21 | 2019-02-12 | 上海常良智能科技有限公司 | Chemical fiber paper tube classification Transmission system based on image recognition |
| CN107328781A (en) * | 2017-05-23 | 2017-11-07 | 镇江苏仪德科技有限公司 | A kind of columnar product detection method of surface flaw and device based on machine vision |
| CN109454006A (en) * | 2018-10-08 | 2019-03-12 | 杭州慧知连科技有限公司 | Chemical fiber wire ingot is stumbled the device and its detection stage division of a defect on-line checking and classification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN204718967U (en) | The accurate disposal system of image in a kind of chemical fiber paper tube outward appearance testing process | |
| CN109507192A (en) | A kind of magnetic core detection method of surface flaw based on machine vision | |
| CN106599890B (en) | digital instrument recognition algorithm for substation inspection robot | |
| CN103051872B (en) | Based on the belt deflection detection method of Edge extraction | |
| CN109879005A (en) | Device for detecting belt tearing and method | |
| CN102680481B (en) | Detection method for cotton fiber impurities | |
| CN111311542A (en) | Product quality detection method and device | |
| Schmugge et al. | Crack segmentation by leveraging multiple frames of varying illumination | |
| CN103759758A (en) | Method for detecting position of automobile meter needle based on mechanical angle and scale identification | |
| CN108802041B (en) | Method for rapidly changing small sample set of screen detection | |
| CN205538710U (en) | Inductance quality automatic check out system based on machine vision | |
| CN106556611A (en) | Chromatograp defect automatic checkout system and detection method based on machine vision | |
| CN104483320A (en) | Digitized defect detection device and detection method of industrial denitration catalyst | |
| CN106093055B (en) | Quality detection method for plural films | |
| CN110852373A (en) | Defect-free sample deep learning network training method based on vision | |
| CN108507484B (en) | Bundled round steel multi-vision visual identifying system and method for counting | |
| CN108871185B (en) | Method, device and equipment for detecting parts and computer readable storage medium | |
| CN110728269B (en) | High-speed rail contact net support pole number plate identification method based on C2 detection data | |
| CN114663744A (en) | Instrument automatic identification method and system based on machine learning | |
| CN110020691A (en) | LCD screen defect inspection method based on the training of convolutional neural networks confrontation type | |
| CN104897688A (en) | Image accurate processing method and system for chemical fiber paper tube appearance detection | |
| CN105976001A (en) | False-proofing license plate making QR code and serial number identifier identification method | |
| CN110688979A (en) | Illegal vehicle tracking method and device | |
| CN105699386B (en) | A kind of automatic cloth inspection labeling method using contact-type image sensor | |
| CN105738376B (en) | A kind of automatic cloth inspecting machine using contact-type image sensor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| AV01 | Patent right actively abandoned |
Granted publication date: 20151021 Effective date of abandoning: 20170825 |
|
| AV01 | Patent right actively abandoned |