CN109900302A - A kind of more groove level tube bubbles error vision measuring method placed in the middle - Google Patents
A kind of more groove level tube bubbles error vision measuring method placed in the middle Download PDFInfo
- Publication number
- CN109900302A CN109900302A CN201910278568.2A CN201910278568A CN109900302A CN 109900302 A CN109900302 A CN 109900302A CN 201910278568 A CN201910278568 A CN 201910278568A CN 109900302 A CN109900302 A CN 109900302A
- Authority
- CN
- China
- Prior art keywords
- bubble
- groove
- area
- interest
- image
- 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
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a kind of more groove level tube bubble error vision measuring methods placed in the middle.Bubble and groove area-of-interest are selected in former levelling tube image center first；Then binaryzation is carried out to the region of interest area image and noise-removed filtering is handled；Secondly each groove is from left to right searched out in region of interest area image after treatment, determine the dependent coordinate information of most inboard two groove and the center abscissa of two grooves is calculated；Then it constructs bubble area-of-interest and intercepts out bubble area-of-interest image from bubble and groove area-of-interest bianry image, it carries out negating processing, interference region is filtered out using morphology opening operation, obtains the maximum connected region of area, as bubble corresponding region；Finally according to bubble center abscissa and most, inboard two groove centers abscissa calculates bubble error amount placed in the middle.More groove level tube bubble errors placed in the middle are measured by using visible sensation method, measurement efficiency and accuracy rate is greatly improved, saves a large amount of human costs.
Description
Technical field
The present invention relates to a kind of machine vision metrology methods used in field of industrial production, more particularly to a kind of more quarters
Line level tube bubble error vision measuring method placed in the middle.
Background technique
Horizontal ruler is the principle using liquid level, directly displays angular displacement with groove in levelling tube and bubble, measures quilt
Side surface relative horizontal position, plumb position, a kind of measurement instrument for tilting departure degree.More groove levelling tubes are in horizontal ruler
An important component, bubble error placed in the middle is to measure an important parameter of horizontal ruler product quality, which leads at present
It to be carried out by the method for manual measurement, it cannot be guaranteed that the accuracy of measurement result, and measurement efficiency is low.With labour at
This rapid increase, traditional manual measurement mode are unable to satisfy the extensive and efficient need that horizontal ruler produces in current industrial
It asks.
Computer vision provides more groove level tube bubbles error survey placed in the middle of a kind of efficient, low cost, high-accuracy
Amount method measures more groove level tube bubble errors placed in the middle using image processing techniques, can not only reduce labor cost, and
And the measurement result of higher precision can be obtained, so that more groove levelling tube testing result confidence levels and detection efficiency will all obtain
It is promoted.But it is up to the present, non-to the research achievement of the vision measuring method of more groove level tube bubble errors placed in the middle both at home and abroad
It is often few.
Based on this, this case proposes a kind of more groove level tube bubbles error vision measuring method placed in the middle.
Summary of the invention
It is carved based on computer vision technique to the purpose of the present invention is technical problem to be solved is just to provide one kind more
The method that line level tube bubble error placed in the middle carries out rapid survey.
In order to achieve the above objectives, the present invention adopts the following technical scheme:
A kind of more groove level tube bubbles error vision measuring method placed in the middle, this method include reading original more groove levels
Pipe image, the acquisition of bubble and groove area-of-interest, the processing of region of interest area image, the inspection of most inboard two scribe line positions
Survey, the acquisition of bubble area-of-interest, the processing of bubble region of interest area image and bubble error calculation placed in the middle and etc..
More groove levelling tube images for processing are from more groove level tube bubbles error vision measurement system placed in the middle
Utilize the real-time acquired image of industrial camera or the prior image data for acquiring and saving.
The acquisition methods of bubble and groove area-of-interest are: selecting one in former levelling tube image center and include all quarters
The rectangular area of line and bubble, it is ensured that this rectangular area bottom is intersected with all grooves, this rectangular area is bubble and quarter
Line area-of-interest.
The processing method of region of interest area image is: being extracted using image partition method to from former levelling tube image
Bubble and groove region of interest area image carry out Threshold segmentation processing, then carry out noise abatement processing using denoising algorithm.
The detection method of most inboard two scribe line positions is: two-value bubble and groove area-of-interest after filtering processing
In image, each groove is searched out by sequence from left to right, determines two most inboard grooves, and recorded the most inboard left side and carve
The position of center line abscissa of line, the right abscissa of most inboard left side groove, the position of center line of most inboard the right groove are horizontal
The left side abscissa of coordinate and most inboard the right groove, according to the position of center line abscissa meter of most inboard two grooves in left and right
Calculate the center abscissa of this two grooves.
The acquisition methods of bubble area-of-interest are: one rectangular area of building, the left side abscissa of this rectangle be most in
The right abscissa of side left side groove, the right abscissa of this rectangle are the left side abscissa of most inboard the right groove, this rectangle
Upper and lower both sides be overlapped respectively with the upper and lower both sides of bubble and groove area-of-interest, this rectangular area is that bubble is interested
Region；The corresponding part of bubble area-of-interest is cut from the two-value bubble and groove region of interest area image after filtering processing
It takes out, obtains bubble region of interest area image.
The processing of bubble region of interest area image specifically: bubble region of interest area image is carried out negating processing, then
Interference region is filtered out using mathematical morphology open operator, obtains the maximum connected region of area, this region is corresponding to bubble
Region.
Bubble error calculation method placed in the middle is: calculating the horizontal seat of bubble center according to the abscissa of bubble area right boundary
Mark sits bubble center abscissa with most inboard two groove center abscissas unification to bubble and groove region of interest area image
Under mark, bubble center abscissa is subtracted with the groove center abscissa after uniform coordinate, difference is that the bubble of image space occupies
Middle error amount；With image calibration value, i.e. actual physical size represented by an image pixel is placed in the middle with the bubble of image space
Error amount is multiplied, that is, obtains practical bubble error amount placed in the middle.
The beneficial effects of the present invention are: a kind of more groove level tube bubble error vision measurements placed in the middle disclosed in this invention
Method obtains levelling tube image by CCD camera in real time, carries out bubble based on image of the image processing techniques to acquisition and misses between two parties
Difference measurements, to solve the problems, such as that efficiency of manual measurement is low, precision is low.Conjunction applied to industrial more groove levelling tube horizontal rulers
Lattice detection, is greatly improved detection efficiency.
Detailed description of the invention
Fig. 1 is error measure overall flow figure placed in the middle of the invention.
Fig. 2 is the levelling tube original image that the present invention obtains.
Fig. 3 is the bubble and groove area-of-interest that the present invention obtains.
Fig. 4 is the bubble and groove region of interest area image that the present invention obtains
Fig. 5 is bubble and groove region of interest area image after the processing that the present invention obtains.
Fig. 6 is the bubble region of interest area image that the present invention obtains.
Fig. 7 be the present invention obtain result is negated to bubble region of interest area image.
Fig. 8 is bubble region of interest area image after the processing that the present invention obtains.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that described herein, specific examples are only used to explain the present invention, not
For limiting the present invention.
Embodiment please refers to Fig. 1 to Fig. 8: Fig. 1 and schematically illustrates more groove level tube bubbles disclosed in this invention
Error measure overall flow placed in the middle；Fig. 2 gives the original levelling tube image that the present invention obtains；Fig. 3 gives acquisition of the present invention
Bubble and groove area-of-interest；Fig. 4 gives the bubble and groove region of interest area image that the present invention obtains；Fig. 5 is provided
Bubble and groove region of interest area image after the processing of the invention obtained；It is interested that Fig. 6 gives the bubble that the present invention obtains
Area image；Fig. 7 give the present invention obtain result is negated to bubble region of interest area image；Fig. 8 gives the present invention and obtains
Bubble region of interest area image after the processing taken.
According to process shown in Fig. 1, more groove level tube bubbles error placed in the middle is measured according to following steps:
(1) the more groove air level images of horizontal ruler are acquired in real time or are read by industrial camera acquire and save in advance
More groove air level image datas obtain levelling tube original-gray image Isrc, as shown in Figure 2.At air level groove in image
In vertical position, it is guaranteed by the tooling of fixed horizontal ruler.
(2) in order to improve measuring speed, the square comprising all grooves and bubble is selected in former levelling tube image center
Shape region, it is ensured that this rectangle bottom is intersected with all grooves, this rectangular area is bubble and groove area-of-interest, is such as schemed
Shown in 3.The area image is intercepted in original image, obtains bubble and groove region of interest area image Iroi, as shown in Figure 4.
(3) in bubble and groove region of interest area image IroiIn, groove and bubble are divided from background using dividing method
It separates out and.Dividing method can use automatic threshold method or fixed threshold method.In the present implementation, using Otsu method to image
IroiAutomatic threshold segmentation is carried out, bianry image I is obtainedb.Due in bianry image IbIn there is likely to be some noise spots, to keep away
Exempt to interfere subsequent processing, needs to be removed interference noise point using Denoising Algorithm, use intermediate value in the present implementation
Filtering carries out noise abatement and handles to obtain image Ifilt, as shown in Figure 5.
(4) since all air level grooves pass through the tooling in more groove level tube bubbles error vision measurement system placed in the middle
Make it all in plumbness, therefore two-value bubble and groove region of interest area image I after filtering processingfiltMost next line
In, it is searched out by sequence from left to right and meets the continuous black line segment of one fixed width to get to each groove.Groove number is even number
N (n accepted value is 2,4,6), determines two most inboard grooves, i.e., the n-th/2 and n-th/2+1 root groove from left to right, and remember
The position of center line abscissa x of most inboard left groove under recordL, most inboard left groove the right abscissa xLR, most inboard right groove
Position of center line abscissa xRAnd the left side abscissa x of most inboard right grooveRL, according in most inboard two grooves in left and right
Heart line position abscissa xLAnd xRCalculate the center abscissa x of this two groovess, it may be assumed that
xs=(xL+xR)/2
(5) in order to obtain the region of interest area image of bubble, in image IfiltOne rectangular area of middle building.This rectangle
Left side abscissa is the right abscissa x of most inboard left grooveLR, the right abscissa of this rectangle is the left side of most inboard right groove
Abscissa xRL, the upper and lower both sides of this rectangle are overlapped with the upper and lower both sides of bubble and groove area-of-interest respectively, this rectangle region
Domain is bubble area-of-interest.From the two-value bubble and groove region of interest area image I after filtering processingfiltIt is middle by bubble sense
The corresponding portion intercepts in interest region come out, and obtain bubble region of interest area image IRoi, as shown in Figure 6.
(6) in order to by bubble from image IRoiIn separate, first to bubble region of interest area image IRoiIt is negated
Processing, then the result that negates that treated using mathematical morphology open operator as shown in fig. 7, filtered out after interference region obtains processing
Bubble region of interest area image Ires, as shown in Figure 8.Finally in image IresMiddle carry out contour detecting obtains the maximum company of area
Logical region, this region is region corresponding to bubble.
(7) according to the abscissa x of bubble area boundary rectangle right boundarybLAnd xbRCalculate bubble center abscissa xb:
xb=(xbL+xbR)/2
By bubble center abscissa xbMost inboard two groove centers abscissa xsUnder the unified levelling tube image coordinate to original
Respectively xBAnd xk:
xB=xb+xRoi
xk=xs
Wherein, xRoiFor bubble region of interest area image IRoiThe upper left corner is in image IroiIn coordinate abscissa value.
With groove center abscissa xkSubtract bubble center abscissa xBThe bubble error amount d placed in the middle of image space can be obtainedI
(unit pixel), it may be assumed that
dI=xk-xB
Further according to image calibration value b (unit mm/pixel), i.e. actual physical size represented by an image pixel,
With the bubble error amount d placed in the middle of image spaceIPractical bubble error amount d placed in the middle can be obtained in multiplicationR(unit mm), it may be assumed that
dR=dI×b
Above-mentioned, although the foregoing specific embodiments of the present invention is described with reference to the accompanying drawings, not protects model to the present invention
The limitation enclosed, those skilled in the art should understand that, based on the technical solutions of the present invention, those skilled in the art are not
Need to make the creative labor the various modifications or changes that can be made still within protection scope of the present invention.
Claims (8)
1. a kind of more groove level tube bubbles error vision measuring method placed in the middle, it is characterised in that: this method includes that reading is original
More groove levelling tube images, the acquisition of bubble and groove area-of-interest, the processing of region of interest area image, most inboard two quarters
The detection of line position, the acquisition of bubble area-of-interest, the processing of bubble region of interest area image and bubble error calculation placed in the middle
And etc..
2. according to claim 1 read original more groove levelling tube images, it is characterised in that: more grooves for processing
Levelling tube image is collected in real time using industrial camera from more groove level tube bubbles error vision measurement system placed in the middle
Image or the image data for acquiring and saving in advance.
3. the acquisition of bubble according to claim 1 and groove area-of-interest, it is characterised in that: in former levelling tube image
Center selects the rectangular area comprising all grooves and bubble, it is ensured that and this rectangular area bottom is intersected with all grooves,
This rectangular area is bubble and groove area-of-interest.
4. the processing of region of interest area image according to claim 1, it is characterised in that: using image partition method to from
The bubble and groove region of interest area image extracted in former levelling tube image carries out Threshold segmentation processing, then using at denoising
Adjustment method carries out noise abatement processing.
5. the detection of most inboard two scribe line positions according to claim 1, it is characterised in that: two after filtering processing
It is worth in bubble and groove region of interest area image, searches out each groove by sequence from left to right, determines two most inboard quarters
Line, and record the right abscissa, most inboard of the position of center line abscissa of most inboard left side groove, most inboard left side groove
The left side abscissa of the position of center line abscissa of the right groove and most inboard the right groove, according to most inboard two quarters of left and right
The position of center line abscissa of line calculates the center abscissa of this two grooves.
6. the acquisition of bubble area-of-interest according to claim 1, it is characterised in that: one rectangular area of building, this
The left side abscissa of rectangle is the right abscissa of most inboard left side groove, and the right abscissa of this rectangle is that most inboard the right is carved
The left side abscissa of line, the upper and lower both sides of this rectangle are overlapped with the upper and lower both sides of bubble and groove area-of-interest respectively, this
Rectangular area is bubble area-of-interest；By bubble from the two-value bubble and groove region of interest area image after filtering processing
The corresponding portion intercepts of area-of-interest come out, and obtain bubble region of interest area image.
7. the processing of bubble region of interest area image according to claim 1, it is characterised in that: to bubble area-of-interest
Image carries out negating processing, then filters out interference region using mathematical morphology open operator, obtains the maximum connected region of area,
This region is region corresponding to bubble.
8. bubble error calculation placed in the middle according to claim 1, it is characterised in that: according to the cross of bubble area right boundary
Coordinate calculates bubble center abscissa, and by bubble center abscissa and most inboard two groove center abscissas are unified to bubble
And under groove area-of-interest image coordinate, bubble center abscissa is subtracted with the groove center abscissa after uniform coordinate,
Difference is the bubble error amount placed in the middle of image space；With image calibration value, i.e. actual physics ruler represented by an image pixel
It is very little, it is multiplied with the bubble of image space error amount placed in the middle, that is, obtains practical bubble error amount placed in the middle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910278568.2A CN109900302A (en) | 2019-04-09 | 2019-04-09 | A kind of more groove level tube bubbles error vision measuring method placed in the middle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910278568.2A CN109900302A (en) | 2019-04-09 | 2019-04-09 | A kind of more groove level tube bubbles error vision measuring method placed in the middle |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109900302A true CN109900302A (en) | 2019-06-18 |
Family
ID=66955509
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910278568.2A Pending CN109900302A (en) | 2019-04-09 | 2019-04-09 | A kind of more groove level tube bubbles error vision measuring method placed in the middle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109900302A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111862055A (en) * | 2020-07-23 | 2020-10-30 | 中国计量大学 | Leveling bubble turning cavity defect detection method based on image processing |
| CN113008267A (en) * | 2021-01-05 | 2021-06-22 | 上海大学 | System and method for detecting qualified limit inclination of level bar based on image processing technology |
| CN113124898A (en) * | 2021-03-03 | 2021-07-16 | 上海大学 | Pipe level device position precision measuring method based on image processing |
| CN117516591A (en) * | 2023-11-07 | 2024-02-06 | 石家庄顾得芯科技有限公司 | Bubble batch performance detection method |
| CN117516591B (en) * | 2023-11-07 | 2024-06-04 | 石家庄顾得芯科技有限公司 | Bubble batch performance detection method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2431990A (en) * | 2005-11-08 | 2007-05-09 | Neil Cottingham | Illuminated spirit level |
| CN108592948A (en) * | 2018-04-28 | 2018-09-28 | 中国计量大学 | A kind of pipe bubble offset method for automatic measurement |
| CN108613630A (en) * | 2018-04-28 | 2018-10-02 | 中国计量大学 | The two linear pipe bubbles based on image processing techniques deviate measuring method |
| CN108876860A (en) * | 2018-04-28 | 2018-11-23 | 中国计量大学 | A kind of image calibration method for pipe bubble offset measurement |
| CN208419963U (en) * | 2018-06-07 | 2019-01-22 | 浙江奇剑工具有限公司 | A kind of horizontal ruler tube bubble |
-
2019
- 2019-04-09 CN CN201910278568.2A patent/CN109900302A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| GB2431990A (en) * | 2005-11-08 | 2007-05-09 | Neil Cottingham | Illuminated spirit level |
| CN108592948A (en) * | 2018-04-28 | 2018-09-28 | 中国计量大学 | A kind of pipe bubble offset method for automatic measurement |
| CN108613630A (en) * | 2018-04-28 | 2018-10-02 | 中国计量大学 | The two linear pipe bubbles based on image processing techniques deviate measuring method |
| CN108876860A (en) * | 2018-04-28 | 2018-11-23 | 中国计量大学 | A kind of image calibration method for pipe bubble offset measurement |
| CN208419963U (en) * | 2018-06-07 | 2019-01-22 | 浙江奇剑工具有限公司 | A kind of horizontal ruler tube bubble |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111862055A (en) * | 2020-07-23 | 2020-10-30 | 中国计量大学 | Leveling bubble turning cavity defect detection method based on image processing |
| CN111862055B (en) * | 2020-07-23 | 2024-03-05 | 中国计量大学 | Bubble turning cavity defect detection method based on image processing |
| CN113008267A (en) * | 2021-01-05 | 2021-06-22 | 上海大学 | System and method for detecting qualified limit inclination of level bar based on image processing technology |
| CN113008267B (en) * | 2021-01-05 | 2023-09-15 | 上海大学 | Level bar limit inclination qualification detection system and method based on image processing technology |
| CN113124898A (en) * | 2021-03-03 | 2021-07-16 | 上海大学 | Pipe level device position precision measuring method based on image processing |
| CN117516591A (en) * | 2023-11-07 | 2024-02-06 | 石家庄顾得芯科技有限公司 | Bubble batch performance detection method |
| CN117516591B (en) * | 2023-11-07 | 2024-06-04 | 石家庄顾得芯科技有限公司 | Bubble batch performance detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109003258B (en) | High-precision sub-pixel circular part measuring method | |
| CN110017773B (en) | Package volume measuring method based on machine vision | |
| CN109900302A (en) | A kind of more groove level tube bubbles error vision measuring method placed in the middle | |
| CN106225770B (en) | The multidimensional digitized record recognition methods of tunnel tunnel face geology and system | |
| CN102032875B (en) | Image-processing-based cable sheath thickness measuring method | |
| CN102151149B (en) | Method and system for automatically measuring ultrasound image of fetus | |
| CN109060821A (en) | Tunnel defect detection method and tunnel defect detection device based on laser detection | |
| CN108613630B (en) | Two-wire tube level bubble offset measurement method based on image processing technology | |
| TWI724696B (en) | Workpiece hole measurement method | |
| CN107358628B (en) | Linear array image processing method based on target | |
| CN108876860A (en) | A kind of image calibration method for pipe bubble offset measurement | |
| CN102988052B (en) | Method and system for measuring foot length | |
| CN112991287B (en) | Automatic indentation measurement method based on full convolution neural network | |
| CN107239742A (en) | A kind of gauge pointer scale value calculating method | |
| CN116165216B (en) | Liquid crystal display micro scratch flaw 3D detection method, system and computing equipment | |
| CN108592948A (en) | A kind of pipe bubble offset method for automatic measurement | |
| CN102274051A (en) | Method and system for automatically measuring bladder volume of ultrasonic image | |
| CN110608685A (en) | Object size rapid measurement method based on raspberry pie | |
| CN108534802A (en) | Pipe bubble based on image processing techniques deviates measuring method | |
| CN111623942A (en) | Displacement measurement method for test structure model of unidirectional vibration table | |
| CN111879264A (en) | Flatness measurement and evaluation system based on line structured light | |
| CN109029203A (en) | A kind of semi-automatic measuring dimension of object device based on Digital Image Processing | |
| CN108320799A (en) | Image analysis and recognition method for lateral flow paper strip disease diagnosis | |
| CN112036409A (en) | Reading identification method and device of liquid level meter | |
| CN105491315A (en) | Projector gamma correction method |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190618 |
|
| WD01 | Invention patent application deemed withdrawn after publication |