CN106651802B - Machine vision scolding tin position finding and detection method - Google Patents
Machine vision scolding tin position finding and detection method Download PDFInfo
- Publication number
- CN106651802B CN106651802B CN201611210061.6A CN201611210061A CN106651802B CN 106651802 B CN106651802 B CN 106651802B CN 201611210061 A CN201611210061 A CN 201611210061A CN 106651802 B CN106651802 B CN 106651802B
- Authority
- CN
- China
- Prior art keywords
- image
- scolding tin
- detection
- template
- machine vision
- 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.)
- Expired - Fee Related
Links
- ATJFFYVFTNAWJD-UHFFFAOYSA-N Tin Chemical group [Sn] ATJFFYVFTNAWJD-UHFFFAOYSA-N 0.000 title claims abstract description 36
- 238000001514 detection method Methods 0.000 title claims abstract description 30
- 238000007781 pre-processing Methods 0.000 claims abstract description 7
- 238000012937 correction Methods 0.000 claims abstract description 4
- 230000005764 inhibitory process Effects 0.000 claims abstract description 4
- 238000005259 measurement Methods 0.000 claims description 14
- 238000000034 method Methods 0.000 claims description 14
- 238000007689 inspection Methods 0.000 claims description 2
- 238000013528 artificial neural network Methods 0.000 description 7
- 229910000679 solder Inorganic materials 0.000 description 6
- 239000013598 vector Substances 0.000 description 5
- 230000007547 defect Effects 0.000 description 4
- 230000004807 localization Effects 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 230000007812 deficiency Effects 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000003466 welding Methods 0.000 description 3
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000284 extract Substances 0.000 description 2
- 238000004321 preservation Methods 0.000 description 2
- 239000000047 product Substances 0.000 description 2
- 239000004065 semiconductor Substances 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 206010061619 Deformity Diseases 0.000 description 1
- 238000003335 Production assurance Methods 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000011430 maximum method Methods 0.000 description 1
- 238000004377 microelectronic Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000005476 soldering Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G06T5/80—
-
- G06T5/70—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30152—Solder
Abstract
Machine vision scolding tin position finding and detection method, belongs to field of image detection, has technical point that including image preprocessing, image registration, detects to anchor point scolding tin；Image preprocessing refers mainly to the inhibition that geometric distortion correction and noise spot are carried out to image, reduces the geometric distortion of reference picture and image to be spliced；Image registration refers mainly to extract the match information in reference picture and image to be spliced, and matching is found in the information extracted, completes the alignment between image.
Description
Technical field
The invention belongs to field of image detection, it is related to Image Acquisition and localization method when a kind of pair of scolding tin positioning.
Background technique
Currently, most of producer is all to be based purely on optical principle to examine to the common deficiency encountered in welding production
It surveys, the two dimensional image of element is obtained by CCD camera, carried out by image procossing, image analysis and computer vision methods
Processing obtains the understanding to image, and then realizes the state expression of the identification of object, positioning and object.With semiconductor, chip
The development of equal microelectronic elements industry, chip is more small towards size, and circuit is more complicated, the more powerful direction hair of function
Exhibition.The spacing for mounting pin is smaller and smaller, and required precision is higher and higher, and thus to detection, more stringent requirements are proposed.But such as
Modern existing AOI detection pattern is only merely machine by camera automatically scanning PCB, acquires image, the solder joint and number of test
It is compared according to the qualified parameter in library and checks defect on PCB and defect is shown/marked by display or Automatic Logos
Will comes out, and modifies for maintenance personal, and detection pattern single in this way is only capable of executing the surface inspection of object, but for part edge
Solder joint detection effect it is just ideal not to the utmost, certainly now there are many AOI can accomplish the photography of multi-angle increase for
The Detection capability that IC foot is stuck up, and increase the camera angle of certain shielded elements, to provide more recall rates.AOI is most main
The deficiency wanted is exactly some grayscale or encounters the larger insufficient light of external environmental interference or detected element shade light and shade not
Machine, which can not accurately find matching area, when obvious causes Detection accuracy to reduce or report by mistake.
Korhonen proposed self-organizing feature map (SOFM) neural network in 1981, and the network is mainly by input layer
The two-tier network constituted with competition layer, input layer is for receiving sample, and competition layer completion classifies to input sample, this
Mode has been widely used in decision-making, machine learning, data mining, file access pattern, image segmentation and pattern classification etc.
Field.In these problems, the prior information of few data is available, and user again it is as few as possible to data a possibility that
It carries out it is assumed that so self-organizing feature map neural network algorithm is especially suitable for checking in data point under this restriction
In relationship, more specifically their composed structure can be assessed.
In the manufacturing process of circuit board, occupy as the solder joint for being connected bridge between circuit unit and circuit board very important
Position.The manufacturing process of solder joint has to pass through the programs such as the control of tin amount, positioning and soldering tin binds, and this program is relatively difficult to control
System, therefore in the manufacturing process of circuit board, the technical requirements of welding are relatively high, and the defect that may then occur is also relatively
It is more.If cannot in time by disfigurement discovery and reparation, will to whole system can reliability service have a huge impact.
From the foregoing, it will be observed that scolding tin positioning is an important procedure to production assurance, however, being substantially in scolding tin positioning
It is carried out based on image, it is seen then that in assembly line engineering, how to obtain the acquisition image of the product using as scolding tin
The image of positioning uses, and also seems extremely important.
Summary of the invention
In order to carry out Image Acquisition to PCB on assembly line, and realize detection and localization, the present invention proposes following skill
Art scheme: machine vision scolding tin position finding and detection method, including image preprocessing, image registration, to anchor point scolding tin detect；Image
Pretreatment refers mainly to the inhibition that geometric distortion correction and noise spot are carried out to image, reduces the several of reference picture and image to be spliced
What distorts；Image registration refers mainly to extract the match information in reference picture and image to be spliced, in the letter extracted
Matching is found in breath, completes the alignment between image；Scolding tin detection, application model matching are carried out to the tack weld in measurement range
Method saved standard scolding tin image as template, when operation, by template and a series of image aspects appearance phase on positions
As subset be compared, constantly adjustment initial threshold, quantify selection standard, count respectively each sample image tone (H),
The histogram of saturation degree (S) and brightness (I) plane, while mutual corresponding threshold value therewith is obtained as a result, immediately to multiple samples
Initial threshold be modified, obtain uniform threshold for image binaryzation, standard grayscale correlation operation is as a kind of form
Convolution, be equivalent to convolution kernel for matched template, a template comprising N pixel is multiplied with the N pixel in detection image
Sum immediately, convolution kernel calculate find out measurement image in each pixel value, the maximum position of result be with template most
Similar place；The machine vision scolding tin position finding and detection method, using a kind of machine vision scolding tin position detecting system, to image
It is acquired, the system comprises: industrial camera, ball integral light source, buzzer, light source controller, conveyer belt, industry calculate
Machine, sensor, display, embedded system, image pick-up card；Industrial camera is connect with image pick-up card, image pick-up card with
Industrial computer is connected, and industrial computer is connect with light source controller, and light source controller is connected with ball integral light source, the ball product
Light splitting source is placed in the lower section of industrial camera, and is located at the top of conveyer belt, and the line of described image capture card and computer is drawn
Two-way connects display all the way, and another way is separately connected sensor and embedded system, and the sensor is located at the upper of conveyer belt
Side.
The utility model has the advantages that can be acquired on assembly line to PCB image using this detection system, also, the image can
It is used with being used in the base image of scolding tin positioning.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of system of the present invention.
Specific embodiment
Embodiment 1: a kind of machine vision scolding tin position detecting system, comprising: industrial camera 1, ball integrate light source 2, buzzing
Device 3, light source controller 4, conveyer belt 5, industrial computer 6, sensor 7, display 8, embedded system 9, image pick-up card 10；
Industrial camera 1 is connect with image pick-up card 10, and image pick-up card 10 is connected with industrial computer 6, industrial computer 6 and light source control
Device 4 processed connects, and light source controller 4 is connected with ball integral light source 2, and the ball integral light source 2 is placed in the lower section of industrial camera, and position
In the top of conveyer belt, the line of described image capture card 10 and computer draws two-way, connects display 8, another way all the way
It is separately connected sensor 7 and embedded system 9, the sensor 7 is located at the top of conveyer belt.
In the present embodiment, it is related to a kind of machine vision scolding tin position finding and detection method, carries out figure using said detecting system
As acquisition, including image preprocessing, image registration, to anchor point scolding tin detect；Image preprocessing refers mainly to carry out image several
The inhibition of what distortion correction and noise spot reduces the geometric distortion of reference picture and image to be spliced；Image registration refers mainly to pair
Match information in reference picture and image to be spliced extracts, and finds and matches in the information extracted, between completion image
Alignment；Scolding tin detection is carried out to the tack weld in measurement range, the matched method of application model makees standard scolding tin image
For template preservation, when operation, template is compared with a series of similar subset of image aspects appearance on positions, constantly tune
Whole initial threshold quantifies selection standard, counts each sample image respectively in tone (H), saturation degree (S) and brightness (I) plane
Histogram, while obtaining mutual corresponding threshold value therewith as a result, being modified immediately to the initial threshold of multiple samples, obtain
Uniform threshold is used for image binaryzation, and standard grayscale correlation operation is used to matched template phase as a form of convolution
When in convolution kernel, a template comprising N pixel is multiplied with the N pixel in detection image and sums immediately, and convolution kernel calculating is found out
Measurement image in each pixel value, the maximum position of result is and the most similar place of template.
Embodiment 2: the supplement as position finding and detection method scheme described in embodiment 1: in order to overcome existing AOI cannot
Some grayscale or encounter the larger insufficient light of external environmental interference be detected element shade light and shade it is unobvious when cannot be accurate
The defect or deficiency of positioning are incorporated into Self-organizing Competition artificial neural network based on original algorithm, and this algorithm is dependent on measurement model
The direction of the initial position and measured object enclosed.It is corresponding at the hard function input neural network Ke Helun of two-value type that data will be relied on
Layer carries out the classification and compression of Self-organizing Competition using guideless training method to input data, self organizing neural network
Input pattern indicates that the similitude for comparing different mode can be converted into the distance for comparing two vectors with vector, that is to say, that choosing
The numerical value of both pattern vectors apart is selected as clustering criterion, the clustering criterion that the present embodiment uses is Euclidean minimum distance method.
The size of a proper measurement range that all scolding tin can be detected on the pcb board is determined by study, then
It is exported using this size and limits correct measurement range on the scolding tin for needing to measure, found in the range size that study obtains
The boundary of scolding tin, finds out measurement coordinate origin therewith, and precision can achieve sub-pixel precision.
To in the detection process due to encounter grayscale or encounter the larger insufficient light of external environmental interference or be detected
The picture quality that element shade light and shade acquires when unobvious is too low, is enhanced the quality of image first to improve the matter of picture
Shooting picture contrast is increased and removes fuzzy and noise by amount, and image is regarded as one by Modified geometrical distortion, frequency of use domain method
Kind 2D signal, carries out the signal enhancing based on two-dimensional Fourier transform to it.The measured zone provided by neural network is made
It is repositioned with area maximum method.
According to the solder pad position in the artificial data calibration training sample learnt before positioning, to what is obtained automatically
Primary color threshold value in welding disking area is counted, and carries out two to the image in measurement range using this threshold value in positioning progress
Value carries out pad localization using the image after optimization.
Scolding tin detection is carried out to the tack weld in measurement range, the matched method of application model makees standard scolding tin image
For template preservation, when operation, template is compared with a series of similar subset of image aspects appearance on positions, initial threshold
Value reaches until solder joint most preferably extracts effect out, quantization selection standard counts each sample respectively by manually constantly regulate specification
Image in tone (H), the histogram of saturation degree (S) and brightness (I) plane, while obtain mutual corresponding threshold value therewith as a result,
The initial threshold of multiple samples is modified immediately, obtains uniform threshold for image binaryzation, standard grayscale correlation fortune
A form of convolution can be used as by calculating, and be used to matched template in systems and be equivalent to convolution kernel, common correlation is with more than
Convolution form is identical, and a template comprising N pixel is multiplied with the N pixel in normal picture and sums immediately.Convolution kernel calculating is asked
Out measurement image in each pixel value, as a result maximum position be and the most similar place of template, standardized vector are as follows:
Using the detection image upper left corner as origin, horizontal and vertical direction is respectively X and Y-axis, and the size of search window is wiWith
hj, tack weld coordinate is Si(xi,yi), keep origin relative position constant, w value range is [- min (0, x21),xi-max
(x2,x21+x1)], h value range is [0, yj-(y21+ y1)] assumes that scolding tin anchor point belongs to the pixel number S of threshold range
(x,y),xs=-min (0, w21),xe=ws-max (w2,w21+w1),ys=0, ye=hs-(h21+h1) then has point (xp,yp)
So that S (xp,yp)=maxS (x, y), operation times are (xe-xs) (ye-ys)(w1hi+w2h2)。
But when all white of image or black are that system will reach a maximum value, this maximum value makes the point
No longer similar to template, the standardized vector of relevance function should be changed in this case:
The influence of the linear change of pixel value in image or template will not be received using above-mentioned expression formula result, if surveying
Result can reach peak-peak 1 when amount target and template exactly match, otherwise be 0, and if negative value occurs in similitude, system is to negative value
Automatic abatement is zero, r2The rate that substitution r has evaded open operation is slow, and final result matching point is expressed as percentage.
Score=max (r, 0)2× 100%
The position finding and detection method as described in the examples, greatly improves the precision of scolding tin pad localization in AOI, with
SMT industry technology makes rapid progress, and the size of chip also can be smaller and smaller, and the precision of pin bonding wire is that topic is bound to become urgently
It solves the problems, such as, precision problem when also solving template selection is combined with neural network and mode-matching technique, by this
A little technologies are applied to mention to the fast development of technique in field in semiconductor packages and industrial vision positioning system
It is supported for effective rapidly.
The preferable specific embodiment of the above, only the invention, but the protection scope of the invention is not
It is confined to this, anyone skilled in the art is in the technical scope that the invention discloses, according to the present invention
The technical solution of creation and its inventive concept are subject to equivalent substitution or change, should all cover the invention protection scope it
It is interior.
Claims (1)
1. a kind of machine vision scolding tin position finding and detection method, which is characterized in that including image preprocessing, image registration, to positioning
Point scolding tin detection；Image preprocessing refers mainly to the inhibition that geometric distortion correction and noise spot are carried out to image, reduces reference picture
With the geometric distortion of image to be spliced；Image registration refers mainly to propose the match information in reference picture and image to be spliced
It takes, matching is found in the information extracted, complete the alignment between image；Scolding tin inspection is carried out to the tack weld in measurement range
It surveys, the matched method of application model carries out scolding tin detection to the tack weld in measurement range, using standard scolding tin image as mould
Plate saves, and when operation, template is compared with a series of similar subset of image aspects appearance on positions, constantly at the beginning of adjustment
Beginning threshold value quantifies selection standard, counts each sample image respectively in the histogram of tone, saturation degree and luminance plane, simultaneously
Mutual corresponding threshold value therewith is obtained as a result, being modified immediately to the initial threshold of multiple samples, uniform threshold is obtained and is used for
Image binaryzation, standard grayscale correlation operation are equivalent to convolution kernel as a form of convolution, for matched template, and one
A template comprising N pixel is multiplied with the N pixel in detection image sums immediately, and convolution kernel calculates in the measurement image found out
Each pixel value, the maximum position of result are and the most similar place of template；The machine vision scolding tin position finding and detection method,
Using a kind of machine vision scolding tin position detecting system, image is acquired, the system comprises: industrial camera (1), ball product
Light splitting source (2), buzzer (3), light source controller (4), conveyer belt (5), industrial computer (6), sensor (7), display
(8), embedded system (9), image pick-up card (10)；Industrial camera (1) is connect with image pick-up card (10), image pick-up card
(10) it is connected with industrial computer (6), industrial computer (6) is connect with light source controller (4), light source controller (4) and ball product
Light splitting source (2) is connected, and ball integral light source (2) is placed in the lower section of industrial camera, and is located at the top of conveyer belt, described image
The line of capture card (10) and computer draws two-way, connects display (8) all the way, another way be separately connected sensor (7) and
Embedded system (9), the sensor (7) are located at the top of conveyer belt.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611210061.6A CN106651802B (en) | 2016-12-24 | 2016-12-24 | Machine vision scolding tin position finding and detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611210061.6A CN106651802B (en) | 2016-12-24 | 2016-12-24 | Machine vision scolding tin position finding and detection method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106651802A CN106651802A (en) | 2017-05-10 |
| CN106651802B true CN106651802B (en) | 2019-10-18 |
Family
ID=58827578
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611210061.6A Expired - Fee Related CN106651802B (en) | 2016-12-24 | 2016-12-24 | Machine vision scolding tin position finding and detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106651802B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107545566A (en) * | 2017-07-27 | 2018-01-05 | 深圳市易飞扬通信技术有限公司 | Visible detection method and system |
| CN109389597B (en) * | 2018-10-24 | 2021-04-27 | 四川长虹电器股份有限公司 | Circuit board defect detection system and method on production line |
| CN110865087A (en) * | 2019-11-27 | 2020-03-06 | 科士恩科技(上海)有限公司 | PCBA quality detection method based on artificial intelligence |
| CN111882617A (en) * | 2020-04-23 | 2020-11-03 | 浙江水晶光电科技股份有限公司 | Monocular calibration method and monocular calibration device |
| CN111504194A (en) * | 2020-05-15 | 2020-08-07 | 深圳市振邦智能科技股份有限公司 | Welding spot positioning method applied to welding spot detection AOI |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102914549A (en) * | 2012-09-10 | 2013-02-06 | 中国航天科技集团公司第五研究院第五一三研究所 | Optical image matching detection method aiming at satellite-borne surface exposed printed circuit board (PCB) soldering joint quality |
| CN103645573A (en) * | 2013-11-22 | 2014-03-19 | 大连日佳电子有限公司 | Liquid crystal display (LCD) detection method and system based on machine vision |
| CN104504375A (en) * | 2014-12-18 | 2015-04-08 | 广州视源电子科技股份有限公司 | Method and device for identifying elements of PCB (Printed Circuit Board) |
| CN105959514A (en) * | 2016-04-20 | 2016-09-21 | 河海大学 | Weak target imaging detection device and method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7773220B2 (en) * | 2008-04-02 | 2010-08-10 | International Business Machines Corporation | Method and system for collecting alignment data from coated chips or wafers |
-
2016
- 2016-12-24 CN CN201611210061.6A patent/CN106651802B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102914549A (en) * | 2012-09-10 | 2013-02-06 | 中国航天科技集团公司第五研究院第五一三研究所 | Optical image matching detection method aiming at satellite-borne surface exposed printed circuit board (PCB) soldering joint quality |
| CN103645573A (en) * | 2013-11-22 | 2014-03-19 | 大连日佳电子有限公司 | Liquid crystal display (LCD) detection method and system based on machine vision |
| CN104504375A (en) * | 2014-12-18 | 2015-04-08 | 广州视源电子科技股份有限公司 | Method and device for identifying elements of PCB (Printed Circuit Board) |
| CN105959514A (en) * | 2016-04-20 | 2016-09-21 | 河海大学 | Weak target imaging detection device and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106651802A (en) | 2017-05-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106651802B (en) | Machine vision scolding tin position finding and detection method | |
| CN106501272A (en) | Machine vision scolding tin position detecting system | |
| CN107945184B (en) | Surface-mounted component detection method based on color image segmentation and gradient projection positioning | |
| CN107230203B (en) | Casting defect identification method based on human eye visual attention mechanism | |
| CN106651857B (en) | A kind of printed circuit board patch defect inspection method | |
| CN108982512A (en) | A kind of circuit board detecting system and method based on machine vision | |
| CN102938077A (en) | Online AOI (Automatic Optical Inspection) image retrieval method based on double-threshold binaryzation | |
| CN111507976B (en) | Defect detection method and system based on multi-angle imaging | |
| CN107886131A (en) | One kind is based on convolutional neural networks detection circuit board element polarity method and apparatus | |
| Guo et al. | Research of the machine vision based PCB defect inspection system | |
| WO2017071406A1 (en) | Method and system for detecting pin of gold needle element | |
| CN109447062A (en) | Pointer-type gauges recognition methods based on crusing robot | |
| TWI765442B (en) | Method for defect level determination and computer readable storage medium thereof | |
| CN110146017A (en) | Industrial robot repetitive positioning accuracy measurement method | |
| Wu et al. | The development of automated solder bump inspection using machine vision techniques | |
| CN206470210U (en) | Machine vision scolding tin position detecting system | |
| CN112014407A (en) | Method for detecting surface defects of integrated circuit wafer | |
| Mukhopadhyay et al. | PCB inspection in the context of smart manufacturing | |
| CN206470205U (en) | Machine vision scolding tin assisted detection system | |
| CN115841488A (en) | Hole checking method of PCB (printed Circuit Board) based on computer vision | |
| CN114820439A (en) | PCB bare board defect detection system and method based on AOI | |
| Wang et al. | Assembly defect detection of atomizers based on machine vision | |
| CN111563869B (en) | Stain test method for quality inspection of camera module | |
| CN107122783A (en) | A kind of assembling connecting piece method for quickly identifying based on Corner Detection | |
| JPH0875418A (en) | Automatic recognition method and automatic recognition device for object |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20191018 Termination date: 20201224 |