CN109632825A - A kind of automatic testing method of coil of strip surface abnormalities protrusion - Google Patents
A kind of automatic testing method of coil of strip surface abnormalities protrusion Download PDFInfo
- Publication number
- CN109632825A CN109632825A CN201910049898.4A CN201910049898A CN109632825A CN 109632825 A CN109632825 A CN 109632825A CN 201910049898 A CN201910049898 A CN 201910049898A CN 109632825 A CN109632825 A CN 109632825A
- Authority
- CN
- China
- Prior art keywords
- coil
- protrusion
- point
- testing method
- automatic testing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Landscapes
- Chemical & Material Sciences (AREA)
- Biochemistry (AREA)
- Pathology (AREA)
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Immunology (AREA)
- Analytical Chemistry (AREA)
- General Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of automatic testing method of coil of strip surface abnormalities protrusion, including to coil of strip acquisition surface data, further comprising the steps of: depth map being mapped as 3D point cloud figure, removes outlier and noise spot；Using remaining stable point cloud chart, it is fitted a plane in the 3 d space；Scanning element cloud is arrived the distance in face using point, determines whether protrusion.The present invention proposes a kind of automatic testing method of coil of strip surface abnormalities protrusion, without a large amount of labeled data and interminable training time, save a large amount of human cost and time cost, it ensure that system to the robustness of illumination, simultaneously possible noise spot is eliminated using RANSAC, the stability of system of raising.
Description
Technical field
The present invention relates to the technical field of intelligence manufacture, the automatic detection side of especially a kind of coil of strip surface abnormalities protrusion
Method.
Background technique
On the hot rolling tandem rolling machine production line of iron and steel enterprise, coil of strip Forming Quality will usually be examined batching region
It surveys, and is handled.Since coil of strip production line is under hyperthermia radiation environment, quality inspection personnel can not close-ups.And meat
There are certain subjectivities for eye observation, and it is tired out to be easy to produce vision, are difficult to realize prolonged online accurate detection.
Prevailing quality defect one of of the abnormal protrusion on surface as coil of strip, it is still main by the way of manually checking at present
It is detected.In order to improve working efficiency, intelligence manufacture is realized, the automatic surface abnormalities protrusion for detecting coil of strip causes people's
Extensive concern.
At present there are two ways to automatic detection surface abnormalities protrusion.A kind of method is using image processing techniques, constantly
Change illumination, contrast changes caused by separating illumination change, and becomes by real-time shadow correcting filter separation contrast
Change and surface characteristics, to be detected to protrusion.The technical disadvantages of this method are:, detection knot sensitive to ambient lighting
Fruit is not sufficiently stable.The stabilization of 10 points of method dependence illumination easily causes detection to fail once illumination variation is not sufficiently stable.Separately
A kind of method is, using depth convolutional neural networks, to carry out feature extraction to image, then pass through nerve using depth learning technology
Network carries out prediction judgement.By the learning training to a large amount of positive negative samples, a network structure of good performance is obtained.It is examining
The survey stage completes feature extraction and identification to image using this trained network, detects abnormal protrusion surface.The party
The technical disadvantages of method are: deployment time is longer, needs a large amount of time cost and human cost.Deep learning needs a large amount of artificial
The positive negative sample of mark, will consume a large amount of manpower and material resources, carrys out labeled data.A network knot of good performance in order to obtain simultaneously
Structure needs to carry out prolonged algorithm training, can also consume a large amount of time cost and economic cost.
Application No. is the applications for a patent for invention of CN109100480A to disclose a kind of coil of strip protuberance defect detecting device and side
Method, method include that shell is placed on to the first inspection positions of the coil of strip；Drive spring described by sliding equipment
Housing bottom movement, the spring drive the rotary encoder and tangent displacement sensor to move on the surface of the coil of strip
Make；The rotary encoder sends the displacement signal of institute's displacement sensors, the contact displacement sensing to the controller
Device sends altitude signal to the controller；After the completion of above-mentioned steps, the shell is moved to second by walking mechanism
Position is detected, is repeated the above steps.The disadvantages of the method are as follows being difficult the Stable sliding on section, device is be easy to cause to fall off, drawn
Safety accident, and higher cost are played, versatility is inadequate, is unfavorable for improving industrial production efficiency.
Summary of the invention
In order to solve the above technical problems, the present invention proposes a kind of automatic testing method of coil of strip surface abnormalities protrusion,
Without a large amount of labeled data and interminable training time, a large amount of human cost and time cost are saved, ensure that system pair
The robustness of illumination, while eliminating possible noise spot using RANSAC, the stability of system of raising.
The present invention provides a kind of automatic testing method of coil of strip surface abnormalities protrusion, including to coil of strip acquisition surface data,
The following steps are included:
Step 1: depth map being mapped as 3D point cloud figure, removes outlier and noise spot；
Step 2: utilizing remaining effective point group, acquire plane equation parameter, be fitted a plane in the 3 d space；
Step 3: scanning element cloud, the distance using point to face determine whether to dash forward according to the size relation of distance and threshold value
It rises.
Preferably, the step 1 includes three points in the random optionally 3D point cloud figure, and the number in face is determined using 3 points
It learns principle and is fitted a plane.
In any of the above-described scheme preferably, the step 1 further includes calculating others in the 3D point cloud figure to own
Point arrives the distance of this plane.
In any of the above-described scheme preferably, the step 1 further includes the point that plan range is recorded and is less than threshold value
Number, iteration 500 times, the distance for finding planar point is less than point, the largest number of planes and all points of threshold value, forms 3D
Effective point group.
In any of the above-described scheme preferably, the step 1 further includes to 3D point cloud figure, and iteration carries out RANSAC calculation
Method removes the discrete point and noise spot of 3D point cloud figure.
In any of the above-described scheme preferably, the step 2 includes following sub-step:
Step 21: setting plane equation expression formula, and list the formula of parameter Z；
Step 22: for n point (x in available point cloudi, yi, zi) make parameter S minimum, wherein n >=3, i=0,
1 ..., n-1 }；
Step 23: being decomposed using mathematical tool SVD, solve parameter a0、a1And a2, thus the plane being fitted, wherein
a0It indicates ..., a1It indicates ..., a2It indicates ....
In any of the above-described scheme preferably, the plane equation expression formula is AX+BY+CZ+D=0, wherein A, B, C
It is equation parameter with D, X, Y and Z respectively indicate the x coordinate at three point point cloud midpoints, and there are also z coordinates for y-coordinate.
In any of the above-described scheme preferably, it according to the plane equation expression formula, obtains SettingThen Z=a0X+a1Y+a2。
In any of the above-described scheme preferably, the calculation formula of the parameter S is
It is preferably in any of the above-described scheme, whenMinimum, the parameter S is minimum, that is, meets
Then have
It is converted into matrix form, is obtained
Wherein, dakIt indicates to akIt differentiates, k ∈ (0,1,2).
The invention proposes one kind, the automatic testing method of coil of strip surface abnormalities protrusion, the final fit Planes of the method
Error is within 1mm, to the abnormal protrusion of 3mm or more, can carry out quickly stable detection.
Detailed description of the invention
Fig. 1 is the process of a preferred embodiment of the automatic testing method of coil of strip surface abnormalities protrusion according to the invention
Figure.
Specific embodiment
The present invention is further elaborated with specific embodiment with reference to the accompanying drawing.
Embodiment one
As shown in Figure 1, executing step 100, coil of strip surface is shot using depth camera, to coil of strip acquisition surface number
According to and obtain the coil of strip exterior view.
Step 110 is executed, depth map is mapped as 3D point cloud figure, removes outlier and noise spot.Include: in this step
Three points in the random optionally 3D point cloud figure, the mathematical principle for determining face using 3 points are fitted a plane；Calculate the 3D
Distance of other all the points to this plane in point cloud chart；Number of the plan range less than the point of threshold value, iteration 500 is recorded
Secondary, the distance for finding planar point is less than point, the largest number of planes and all points of threshold value, forms the effective point group of 3D；It is right
3D point cloud figure, iteration carry out RANSAC algorithm, remove the discrete point and noise spot of 3D point cloud figure.
It executes step 120 and acquires plane equation parameter using remaining effective point group, be fitted one in the 3 d space and put down
Face.It in this step include: 1) to set plane equation expression formula AX+BY+CZ+D=0, wherein A, B, C and D are equation parameter,
X, Y and Z respectively indicates the x coordinate at three point point cloud midpoints, and there are also z coordinates for y-coordinate.It obtains
SettingThen Z=a0X+a1Y+a2。
2) for n point (x in available point cloudi, yi, zi) make parameter S minimum, wherein n >=3, i={ 0,1 ..., n-
1 }, the calculation formula of parameter S is
WhenMinimum, the parameter S is minimum, that is, meets
Then have
It is converted into matrix form, is obtained
Wherein, dakIt indicates to akIt differentiates, k ∈ (0,1,2).
3) it is decomposed using mathematical tool SVD, solves parameter a0、a1And a2, thus the plane being fitted.
Step 130 is executed, scanning element cloud judges a little whether be greater than threshold value to the distance in face.If it is less than threshold value, then execute
Step 132, determine that abnormal protrusion is not present in coil of strip surface.If it is larger than or equal to threshold value, 140 are thened follow the steps, determines coil of strip table
There is abnormal protrusion in face.
Embodiment two
In order to solve the problems, such as that effect is unstable and higher cost, method proposes a kind of stabilizations, not by ambient lighting
The automatic detection algorithm of influence and a kind of lower-cost coil of strip surface abnormalities protrusion.The method is protected due to utilizing depth information
It system has been demonstrate,proved to the robustness of illumination, while eliminating possible noise spot using RANSAC, the stability of system of raising.
Simultaneously as the method is that non-statistical learns class method, it is not necessarily to a large amount of labeled data and interminable training time, is saved a large amount of
Human cost and time cost.The final fit Plane error of the method, can to the abnormal protrusion of 3mm or more within 1mm
To carry out quickly stable detection.
Using depth camera carry out coil of strip exception protrusion automatic testing method the following steps are included:
Step 1: first with depth camera universal on the market, to coil of strip acquisition surface data.
Step 2: depth map is mapped as 3D point cloud figure, removes outlier and noise spot.
Using depth information, original depth-map is mapped as 3D point cloud figure by corresponding 2D coordinate, can from the side view of 3D cloud atlas
To find out, there is abnormal protrusion near coil edges.Depth map is converted into two-dimensional matrix form according to graph model.So
Afterwards, three points in random optionally 3D point cloud figure, the mathematical principle for determining face using 3 points are fitted a plane.Calculating 3D point cloud
Middle others all the points record simultaneously to the distance of this plane, the number of the point of threshold value are less than to plan range.Using RANSAC
Algorithm iteration 500 times, find planar point distance be less than threshold value point the largest number of planes and all points, formed 3D
Effective point group.Remove the outlier and noise spot outside effective point group.
The basic assumption of RANSAC algorithm is in sample comprising correct data (inliers, the number that can be described by model
According to), also comprising abnormal data (outliers, deviation normal range (NR) is far, can not adapt to the data of mathematical model), i.e. data set
In contain noise.These abnormal datas may be due to generations such as wrong measurement, the calculating of hypothesis, mistake of mistake.Together
When RANSAC also assume that, give one group of correct data, there is the method that can calculate the model parameter for meeting these data.
The basic thought of RANSAC algorithm describes:
1. considering the model that a minimum sampling cardinality is n (n is smallest sample number needed for initialization model parameter)
With a sample set P, sample number # (P) > n of set P, the subset S initialization mould of the P comprising n sample is randomly selected from P
Type M；
2. complementary set SC=P sample set in S with the error of model M less than a certain given threshold t and S constitute S*.S* recognizes
To be interior point set, they constitute the consistent collection (Consensus Set) of S；
3. if # (S*) >=N, it is believed that obtain correct model parameter, and using collection S* (interior point inliers) using minimum two
The methods of multiply and to recalculate new model M *；Again new S is randomly selected, above procedure is repeated；
4. algorithm fails if not finding consistent collection after completing certain frequency in sampling, obtained after otherwise choosing sampling
The consistent collection of maximum judge in exterior point, algorithm terminates.
Step 3: remaining effective point group is utilized, plane equation parameter is acquired.Using least square method, finds one and put down
Face, so that the sum of the distance of all the points to the plane is minimum.This plane i.e. we according to the effective point group of 3D, the 3D plane of fitting.
Specific steps and principle: the expression formula of plane equation is set are as follows: AX+BY+CZ+D=0, thenWherein, A, B, C and D are equation parameter, and X, Y and Z respectively indicate the x at three point point cloud midpoints
Coordinate, there are also z coordinates for y-coordinate.We enableThen Z=a0X+a1Y+a2.For
N point (n >=3) in available point cloud, (xi, yi, zi), i=0,1 ..., n-1.As long as making
Minimum, then a at this time0, a1, a2Being formed by plane is optimal planar.Keep S minimum, then should meet,Most
It is small, wherein dakIt indicates to akIt differentiates, k ∈ (0,1,2).I.e.
Then have:
It is converted into matrix form, is had:
It is decomposed using Conventional mathematical tool SVD, parameter a can be acquired0, a1, a2, thus the plane being fitted.
This plane i.e. we according to the effective point group of 3D, the 3D plane of fitting.
The basic definition of least square method: small square law (also known as least squares method) is a kind of mathematical optimization techniques.It is logical
Cross the optimal function matching for minimizing the quadratic sum searching data of error.It can easily be acquired using least square method unknown
Data, and the quadratic sum of error is minimum between the data and real data that these are acquired.
Step 4: scanning element cloud, the distance using point to face determine whether to dash forward according to the size relation of distance and threshold value
It rises.
For a better understanding of the present invention, the above combination specific embodiments of the present invention are described in detail, but are not
Limitation of the present invention.Any simple modification made to the above embodiment according to the technical essence of the invention, still belongs to
In the range of technical solution of the present invention.In this specification the highlights of each of the examples are it is different from other embodiments it
Locate, the same or similar part cross-reference between each embodiment.For system embodiments, due to itself and method
Embodiment corresponds to substantially, so being described relatively simple, the relevent part can refer to the partial explaination of embodiments of method.
Claims (10)
1. a kind of automatic testing method of coil of strip surface abnormalities protrusion, including to coil of strip acquisition surface data, which is characterized in that also
The following steps are included:
Step 1: depth map being mapped as 3D point cloud figure, removes outlier and noise spot；
Step 2: utilizing remaining effective point group, acquire plane equation parameter, be fitted a plane in the 3 d space；
Step 3: scanning element cloud, the distance using point to face determine whether protrusion according to the size relation of distance and threshold value.
2. the automatic testing method of coil of strip surface abnormalities protrusion as described in claim 1, it is characterised in that: step 1 packet
Three points in the random optionally 3D point cloud figure are included, the mathematical principle for determining face using 3 points is fitted a plane.
3. the automatic testing method of coil of strip surface abnormalities protrusion as claimed in claim 2, it is characterised in that: the step 1 is also
Distance including other all the points in the calculating 3D point cloud figure to this plane.
4. the automatic testing method of coil of strip surface abnormalities protrusion as claimed in claim 3, it is characterised in that: the step 1 is also
Including be recorded plan range less than threshold value point number, iteration 500 times, find planar point distance be less than threshold value point,
The largest number of planes and all points form the effective point group of 3D.
5. the automatic testing method of coil of strip surface abnormalities protrusion as claimed in claim 4, it is characterised in that: the step 1 is also
Including to 3D point cloud figure, iteration carries out RANSAC algorithm, the discrete point and noise spot of removal 3D point cloud figure.
6. the automatic testing method of coil of strip surface abnormalities protrusion as described in claim 1, it is characterised in that: step 2 packet
Include following sub-step:
Step 21: setting plane equation expression formula, and list the formula of parameter Z；
Step 22: for n point (x in available point cloudi, yi, zi) make parameter S minimum, wherein n >=3, i=(0,1 ...,
n-1}；
Step 23: being decomposed using mathematical tool SVD, solve parameter a0、a1And a2, thus the plane being fitted.
7. the automatic testing method of coil of strip surface abnormalities protrusion as claimed in claim 6, it is characterised in that: the plane equation
Expression formula is AX+BY+CZ+D=0, wherein A, B, C and D are equation parameter, and the x that X, Y and Z respectively indicate three point point cloud midpoints is sat
Mark, there are also z coordinates for y-coordinate.
8. the automatic testing method of coil of strip surface abnormalities protrusion as claimed in claim 7, it is characterised in that: according to the plane
Equation expression formula, obtainsSettingThen Z
=a0X+a1Y+a2。
9. the automatic testing method of coil of strip surface abnormalities protrusion as claimed in claim 8, it is characterised in that: the parameter S's
Calculation formula is
10. the automatic testing method of coil of strip surface abnormalities protrusion as claimed in claim 9, it is characterised in that: whenMinimum,
The parameter S is minimum, that is, meets
Then have
It is converted into matrix form, is obtained
Wherein, dakIt indicates to akIt differentiates, k ∈ (0,1,2).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910049898.4A CN109632825A (en) | 2019-01-18 | 2019-01-18 | A kind of automatic testing method of coil of strip surface abnormalities protrusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910049898.4A CN109632825A (en) | 2019-01-18 | 2019-01-18 | A kind of automatic testing method of coil of strip surface abnormalities protrusion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109632825A true CN109632825A (en) | 2019-04-16 |
Family
ID=66061381
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910049898.4A Pending CN109632825A (en) | 2019-01-18 | 2019-01-18 | A kind of automatic testing method of coil of strip surface abnormalities protrusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109632825A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110428175A (en) * | 2019-08-05 | 2019-11-08 | 东北大学秦皇岛分校 | A kind of Hot Strip Crown Prediction of Media method based on deep learning |
| CN113269759A (en) * | 2021-05-28 | 2021-08-17 | 中冶赛迪重庆信息技术有限公司 | Steel coil information detection method, system, medium and terminal based on image recognition |
| CN113280730A (en) * | 2020-02-19 | 2021-08-20 | 宝钢日铁汽车板有限公司 | System and method for efficiently detecting strip head of steel coil |
| CN114627020A (en) * | 2022-03-18 | 2022-06-14 | 易思维（杭州）科技有限公司 | Method for removing light-reflecting noise points of curved surface workpiece |
| CN117288770A (en) * | 2023-11-24 | 2023-12-26 | 张家港市鸿海精密模具有限公司 | Multi-dimensional detection method and system for surface defects of blow molding die |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150116481A1 (en) * | 2013-10-29 | 2015-04-30 | The Boeing Company | Mapping damaged regions on objects |
| CN105759662A (en) * | 2014-10-13 | 2016-07-13 | 通用电气公司 | Method And Device For Automatically Identifying Point Of Interest On The Surface Of Anomaly |
| CN106813568A (en) * | 2015-11-27 | 2017-06-09 | 阿里巴巴集团控股有限公司 | object measuring method and device |
| CN107064170A (en) * | 2017-04-11 | 2017-08-18 | 深圳市深视智能科技有限公司 | One kind detection phone housing profile tolerance defect method |
| CN107238374A (en) * | 2017-05-04 | 2017-10-10 | 华南农业大学 | A kind of classification of concave plane part and recognition positioning method |
-
2019
- 2019-01-18 CN CN201910049898.4A patent/CN109632825A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150116481A1 (en) * | 2013-10-29 | 2015-04-30 | The Boeing Company | Mapping damaged regions on objects |
| CN105759662A (en) * | 2014-10-13 | 2016-07-13 | 通用电气公司 | Method And Device For Automatically Identifying Point Of Interest On The Surface Of Anomaly |
| CN106813568A (en) * | 2015-11-27 | 2017-06-09 | 阿里巴巴集团控股有限公司 | object measuring method and device |
| CN107064170A (en) * | 2017-04-11 | 2017-08-18 | 深圳市深视智能科技有限公司 | One kind detection phone housing profile tolerance defect method |
| CN107238374A (en) * | 2017-05-04 | 2017-10-10 | 华南农业大学 | A kind of classification of concave plane part and recognition positioning method |
Non-Patent Citations (1)
| Title |
|---|
| 魏英姿等: "基于随机抽取一致性的稳健点云平面拟合", 《北京工业大学学报》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110428175A (en) * | 2019-08-05 | 2019-11-08 | 东北大学秦皇岛分校 | A kind of Hot Strip Crown Prediction of Media method based on deep learning |
| CN110428175B (en) * | 2019-08-05 | 2022-05-24 | 东北大学秦皇岛分校 | Hot rolled strip steel plate convexity prediction method based on deep learning |
| CN113280730A (en) * | 2020-02-19 | 2021-08-20 | 宝钢日铁汽车板有限公司 | System and method for efficiently detecting strip head of steel coil |
| CN113269759A (en) * | 2021-05-28 | 2021-08-17 | 中冶赛迪重庆信息技术有限公司 | Steel coil information detection method, system, medium and terminal based on image recognition |
| CN114627020A (en) * | 2022-03-18 | 2022-06-14 | 易思维（杭州）科技有限公司 | Method for removing light-reflecting noise points of curved surface workpiece |
| CN117288770A (en) * | 2023-11-24 | 2023-12-26 | 张家港市鸿海精密模具有限公司 | Multi-dimensional detection method and system for surface defects of blow molding die |
| CN117288770B (en) * | 2023-11-24 | 2024-02-06 | 张家港市鸿海精密模具有限公司 | Multi-dimensional detection method and system for surface defects of blow molding die |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109632825A (en) | A kind of automatic testing method of coil of strip surface abnormalities protrusion | |
| CN109255813B (en) | Man-machine cooperation oriented hand-held object pose real-time detection method | |
| CN108898047B (en) | Pedestrian detection method and system based on blocking and shielding perception | |
| CN106408591B (en) | A kind of anti-method for tracking target blocked | |
| CN104062973B (en) | A kind of mobile robot based on logos thing identification SLAM method | |
| CN104299247B (en) | A kind of video target tracking method based on adaptive measuring matrix | |
| CN105913013A (en) | Binocular vision face recognition algorithm | |
| CN108596009A (en) | A kind of obstacle detection method and system for agricultural machinery automatic Pilot | |
| CN110648362B (en) | Binocular stereo vision badminton positioning identification and posture calculation method | |
| CN102982334B (en) | The sparse disparities acquisition methods of based target edge feature and grey similarity | |
| CN110263920A (en) | Convolutional neural networks model and its training method and device, method for inspecting and device | |
| CN109345513B (en) | Cigarette package defect detection method with cigarette package posture calculation function | |
| CN105354819A (en) | Depth data measurement system, depth data determination method and apparatus | |
| JP2020161129A (en) | System and method for scoring color candidate poses against color image in vision system | |
| KR101460313B1 (en) | Apparatus and method for robot localization using visual feature and geometric constraints | |
| CN106504262A (en) | A kind of small tiles intelligent locating method of multiple features fusion | |
| CN111126116A (en) | Unmanned ship river channel garbage identification method and system | |
| US9953411B2 (en) | Method for processing a digital image of the surface of a tyre in order to detect an anomaly | |
| CN115512215A (en) | Underwater biological monitoring method and device and storage medium | |
| CN116152697A (en) | Three-dimensional model measuring method and related device for concrete structure cracks | |
| CN108388854A (en) | A kind of localization method based on improvement FAST-SURF algorithms | |
| CN112036269A (en) | Fall detection method and device, computer equipment and storage medium | |
| CN115862124A (en) | Sight estimation method and device, readable storage medium and electronic equipment | |
| CN115063762A (en) | Method, device and equipment for detecting lane line and storage medium | |
| CN111160280A (en) | RGBD camera-based target object identification and positioning method and mobile robot |
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 |
Application publication date: 20190416 |
|
| RJ01 | Rejection of invention patent application after publication |