CN117237434A - H-shaped steel size measurement method and device - Google Patents
H-shaped steel size measurement method and device Download PDFInfo
- Publication number
- CN117237434A CN117237434A CN202311516536.4A CN202311516536A CN117237434A CN 117237434 A CN117237434 A CN 117237434A CN 202311516536 A CN202311516536 A CN 202311516536A CN 117237434 A CN117237434 A CN 117237434A
- Authority
- CN
- China
- Prior art keywords
- light bar
- bar image
- shaped steel
- pixel point
- determining
- 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.)
- Granted
Links
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 152
- 239000010959 steel Substances 0.000 title claims abstract description 152
- 238000000691 measurement method Methods 0.000 title claims description 5
- 238000000034 method Methods 0.000 claims abstract description 70
- 238000000611 regression analysis Methods 0.000 claims abstract description 23
- 238000012216 screening Methods 0.000 claims description 22
- 238000012937 correction Methods 0.000 claims description 16
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000012417 linear regression Methods 0.000 claims description 10
- 230000001678 irradiating effect Effects 0.000 claims description 6
- 238000005259 measurement Methods 0.000 abstract description 33
- 238000000605 extraction Methods 0.000 description 22
- 238000012545 processing Methods 0.000 description 12
- 238000004422 calculation algorithm Methods 0.000 description 11
- 238000004590 computer program Methods 0.000 description 6
- 238000001514 detection method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000001788 irregular Effects 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000009466 transformation Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000001419 dependent effect Effects 0.000 description 2
- 230000010354 integration Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 229910000746 Structural steel Inorganic materials 0.000 description 1
- 230000004075 alteration Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000009440 infrastructure construction Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000009021 linear effect Effects 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention discloses a method and a device for measuring the size of H-shaped steel, relates to the technical field of workpiece size measurement, and aims to solve the problems of low speed and low precision of the conventional H-shaped steel size measurement. The method comprises the following steps: acquiring a light bar image of the H-shaped steel; correcting and extracting a target pixel point set forming a light bar center line in the light bar image based on a local regression analysis method and secondary fitting; extracting first characteristic points meeting preset conditions in the light bar images corresponding to each direction; determining second characteristic points in the first direction light bar image and the second direction light bar image based on the first characteristic points, the target pixel point set and the slopes corresponding to the target pixel point set; and determining the section size of the H-shaped steel according to the first characteristic points and the second characteristic points. The H-shaped steel dimension measuring method provided by the invention is used for improving the accuracy and speed of H-shaped steel dimension measurement.
Description
Technical Field
The invention relates to the technical field of workpiece size measurement, in particular to a method and a device for measuring the size of H-shaped steel.
Background
China is a large country of manufacturing industry, and plays an important role in the field of infrastructure construction. The infrastructure and the building are constructed by using a large amount of structural steel including H-shaped steel, and the H-shaped steel is strongly demanded. In manufacturing, quality inspection of the product is critical. The size of the H-steel is also critical as a key factor affecting the service performance.
At present, the development of the machine vision technology is rapid, and the machine vision technology has the natural advantages of high measurement precision, good real-time performance, non-contact performance and the like, and has become a popular field of research. The visual detection technology is introduced to the measurement of the H-shaped steel size, so that the H-shaped steel size detection efficiency can be greatly improved, and the qualification rate of products can be improved. The monocular camera-single line structured light model in visual detection is applicable to size measurement under the scene due to simple model, obvious characteristics and high measurement precision. The existing measurement method for measuring the H-shaped steel size by adopting the model generally searches straight lines in an image based on Hough transformation so as to solve the H-shaped steel size, however, the nature of the Hough transformation detection straight lines is a voting mechanism, false peaks can be generated in a parameter space by non-straight lines, and then the situation that the found straight lines have multiple solutions or misunderstandings can occur, so that screening complexity is increased, algorithm complexity is high, and operation rate is reduced; secondly, two short horizontal light stripes on the flange thickness can be identified into a straight line during detection, and when the height difference of the flange of the H-shaped steel is large, the measurement size can be greatly influenced.
Disclosure of Invention
The invention aims to provide a method and a device for measuring the dimension of H-shaped steel, which are used for improving the accuracy and the measuring efficiency of the measurement of the cross-section dimension of the H-shaped steel.
In order to achieve the above object, the present invention provides the following technical solutions:
in one aspect, the invention provides a method for measuring the size of H-shaped steel, comprising the following steps: acquiring a light bar image of the H-shaped steel; the light bar image is an image formed by irradiating the H-shaped steel with a laser; the light bar images comprise a first direction light bar image, a second direction light bar image, a third direction light bar image and a fourth direction light bar image;
extracting a target pixel point set forming a light bar center line in the light bar image based on a local regression analysis method and secondary fitting correction;
extracting first characteristic points meeting preset conditions in the light bar image corresponding to each direction;
determining second characteristic points in the first direction light bar image and the second direction light bar image based on the first characteristic points, the target pixel point set and slopes corresponding to the target pixel point set;
and determining the section size of the H-shaped steel according to the first characteristic points and the second characteristic points.
In a second aspect, the invention also provides an H-section steel dimension measuring device, which comprises an H-section steel light bar image acquisition module, a light bar image acquisition module and a light bar image acquisition module, wherein the light bar image acquisition module is used for acquiring the light bar image of the H-section steel; the light bar image is an image formed by irradiating the H-shaped steel with a laser; the light bar images comprise a first direction light bar image, a second direction light bar image, a third direction light bar image and a fourth direction light bar image;
The target pixel point set determining module is used for extracting a target pixel point set forming a light bar center line in the light bar image based on a local regression analysis method and secondary fitting correction;
the first characteristic point determining module is used for extracting first characteristic points meeting preset conditions in the light bar image corresponding to each direction;
the second characteristic point determining module is used for determining second characteristic points in the first direction light bar image and the second direction light bar image based on the first characteristic points, the target pixel point set and slopes corresponding to the target pixel point set;
and the section size calculation module is used for determining the section size of the H-shaped steel according to the first characteristic points and the second characteristic points.
The beneficial effects are that: compared with the prior art, in the H-shaped steel dimension measuring method provided by the invention, the light bar image of the H-shaped steel is obtained; the measurement scheme of the invention adopts four independent cameras and laser models to measure the cross section size of the H-shaped steel, the cross section size is not interfered with each other during working, errors caused by cooperative work are not generated, the overall measurement accuracy is high, and the target pixel point set for forming the center line of the light bar in the extracted light bar image is corrected based on a local regression analysis method and secondary fitting; the partial regression analysis can reduce the operand, improve the extraction speed of the target pixel point set of the light bar center line, and perform twice fitting correction on the light bar center line by twice fitting correction, so that the light bar center line is smoother, and the extracted target pixel point set of the light bar center line is more accurate; extracting first characteristic points meeting preset conditions in the light bar images corresponding to each direction; determining second characteristic points in the first direction light bar image and the second direction light bar image based on the first characteristic points, the target pixel point set and the slopes corresponding to the target pixel point set; and determining the section size of the H-shaped steel according to the first characteristic points and the second characteristic points. The algorithm of the local regression analysis and the secondary fitting correction of the target pixel point set of the light bar center optimizes the light bar center line extraction precision aiming at the irregular light bar shape of the H-shaped steel surface, and can output the normal line and tangential direction of each target pixel point, and some characteristic points in the section size of the H-shaped steel can be extracted by taking the normal line and tangential direction as constraint conditions, so that the high-precision real-time measurement of the section size of the H-shaped steel is realized.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:
FIG. 1 is a flow chart of a method for measuring the dimension of H-shaped steel provided by the invention;
FIG. 2 is a schematic view of the placement positions of a laser and a camera provided by the present invention;
FIG. 3 is a view of a first directional light bar image acquired by an upper camera according to the present invention;
FIG. 4 is a fourth view of a light bar image acquired by a right camera according to the present invention;
FIG. 5 is a schematic diagram of a first feature point and a second feature point according to the present invention;
FIG. 6 is a schematic cross-sectional dimension of H-section steel to be measured according to the present invention;
FIG. 7 is an image of an H-beam light bar with a broken line at room temperature provided by the invention;
FIG. 8 is an image of an H-beam light bar with a broken wire at high temperature provided by the invention;
FIG. 9 is a schematic view of an image of a region of interest (ROI) comprising a light bar according to the present invention;
FIG. 10 is a flowchart for extracting coordinates of a target pixel point of a light bar center line according to the present invention;
FIG. 11 is a schematic diagram of a target pixel point integration method of a light bar according to the present invention;
FIG. 12 is a schematic diagram showing a target pixel point integration method line direction neighborhood of a light bar according to the present invention;
FIG. 13 is a schematic view of a normal direction of a center line of a light bar according to the present invention;
FIG. 14 is a schematic view of a parabolic fit of a normal-direction cross-section provided by the present invention;
FIG. 15 is a schematic view of a second linear effect provided by the present invention;
FIG. 16 is a schematic view of an ideal H-flange thickness roof model provided by the present invention;
fig. 17 is a schematic structural diagram of an H-section steel dimension measuring device according to the present invention.
Reference numerals:
21-H-shaped steel, 22-laser and 23-camera.
Detailed Description
In order to clearly describe the technical solution of the embodiments of the present invention, in the embodiments of the present invention, the words "first", "second", etc. are used to distinguish the same item or similar items having substantially the same function and effect. For example, the first threshold and the second threshold are merely for distinguishing between different thresholds, and are not limited in order. It will be appreciated by those of skill in the art that the words "first," "second," and the like do not limit the amount and order of execution, and that the words "first," "second," and the like do not necessarily differ.
In the present invention, the words "exemplary" or "such as" are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary" or "such as" is intended to present related concepts in a concrete fashion.
In the present invention, "at least one" means one or more, and "a plurality" means two or more. "and/or", describes an association relationship of an association object, and indicates that there may be three relationships, for example, a and/or B, and may indicate: a alone, a and B together, and B alone, wherein a, B may be singular or plural. The character "/" generally indicates that the context-dependent object is an "or" relationship. "at least one of" or the like means any combination of these items, including any combination of single item(s) or plural items(s). For example, at least one (one) of a, b or c may represent: a, b, c, a and b, a and c, b and c, or a, b and c, wherein a, b, c can be single or multiple.
In the existing method for measuring the section size of the H-shaped steel, a steger algorithm is adopted for extracting the center line of the light bar, the steger algorithm is an algorithm for detecting the edge of an image, the Heisen matrix operation in the method needs to carry out convolution operation on the image for 5 times, so that the algorithm complexity is high, the operation rate is reduced, and in addition, the extraction precision of the method is poor, so that the precision of a final measurement result is low.
In order to solve the problems, the invention provides a method and a device for measuring the dimension of H-shaped steel, which adopt four independent cameras and laser models to measure the dimension of the section of the H-shaped steel, improve the measuring precision and the measuring speed, and are described with reference to the accompanying drawings.
FIG. 1 is a flow chart of a method for measuring the dimension of H-shaped steel, which is shown in FIG. 1, and comprises the following steps:
step 101, acquiring a light bar image of H-shaped steel;
the method is characterized in that the section size of the H-shaped steel is measured by adopting a monocular camera and a single-line structure light model, wherein each monocular camera and the single-line structure light model are composed of a laser and a camera, a light bar image is formed by irradiating the laser on the H-shaped steel, the light bar image is collected through the camera, according to the shape characteristics of the H-shaped steel, the placement positions of the laser and the camera are shown in fig. 2, the H-shaped steel 21 comprises flanges and webs, when the H-shaped section of the H-shaped steel 21 to be measured is positive, one laser 22 is respectively placed in four preset directions such as right above, right below and right left Fang Yiji of the H-shaped steel 21, laser emitted by the laser 22 is perpendicularly irradiated on the surface of the H-shaped steel 21, a camera 23 is correspondingly placed beside each laser 22, and the included angle between the camera 23 and the laser 22 is a preset included angle. Meanwhile, the cameras 23 are horizontally arranged relative to the light bars, the light bars collected in the left and right directions are identical in shape and characteristics, and the light bars are conveniently extracted.
After the measurement platform composed of four sets of monocular cameras and the single-line structure optical model is built, vector cross multiplication of Zhang Zhengyou is adopted to respectively and independently calibrate the four sets of monocular cameras and the single-line structure optical model in the measurement platform, camera external parameters of each camera are obtained, and after calibration is completed, high-precision gauge blocks are adopted to verify calibration precision so as to ensure the calibration precision. After calibration, the light bar images are collected through the cameras, wherein the light bar images comprise first-direction light bar images collected by the cameras corresponding to the upper part, second-direction light bar images collected by the cameras corresponding to the lower part, third-direction light bar images collected by the cameras corresponding to the left part, and fourth-direction light bar images collected by the cameras corresponding to the right part. As shown in fig. 3, which is a first-direction light bar image collected, and as shown in fig. 4, which is a fourth-direction light bar image collected, the first-direction light bar image and the second-direction light bar image have the same shape, and the opening directions are opposite, so that the method used for extracting the feature points is the same, the third-direction light bar image and the fourth-direction light bar image have the same shape, and the method used for extracting the feature points is the same, and therefore, only the first-direction light bar image and the fourth-direction light bar image are described as examples. The camera external parameters are used for converting the first feature point and the second feature point into world coordinates, namely actual coordinates.
Step 102: extracting a target pixel point set forming a light bar center line in the light bar image based on a local regression analysis method and secondary fitting correction;
the local regression analysis method is to take the pixel point of the central line of the light bar extracted in a rough way as a central frame to select a neighborhood with a preset neighborhood radius, screen the pixel points meeting preset pixel conditions in the neighborhood selected by the frame, and fit the trend of the pixel points in the field for the first time through a linear regression formula to obtain the normal direction corresponding to the pixel points of the central line of the light bar extracted in the rough way. And then performing a second parabolic fitting in the neighborhood selected by the frame to obtain an offset, finishing the second fitting correction, and finally determining a target pixel point set of the light bar center line according to the offset and the pixel points of the light bar center line which are roughly extracted.
Step 103: extracting first characteristic points meeting preset conditions in the light bar image corresponding to each direction;
referring to fig. 5, the first feature points include: the first edge points of the flange, such as point 1 and point 2, and the edge points of the side surface of the H-section, such as point 11 and point 12, wherein when the H-section is the front surface, the side surface of the H-section is the front left surface and the front right surface, and it should be noted that fig. 5 only shows the first feature points corresponding to the first direction light bar image and the fourth direction light bar image, and the first feature points also include the first edge points of the flange of the second direction light bar image and the edge points of the side surface of the H-section of the third direction light bar image.
Step 104: determining second characteristic points in the first direction light bar image and the second direction light bar image based on the first characteristic points, the target pixel point set and slopes corresponding to the target pixel point set;
referring to fig. 5, the second feature point includes a flange center point: as with points 3 and 4, the flange second boundary points: as for points 5 and 6, four characteristic points of the web: point 7, point 8, point 9, point 10. It should be noted that fig. 5 only shows the second feature points corresponding to the first direction light bar image, and the second feature points further include four feature points of the flange center point, the flange second boundary point, and the web of the second direction light bar image.
Step 105: and determining the section size of the H-shaped steel according to the first characteristic points and the second characteristic points.
The cross-sectional dimensions of the H-section steel, see fig. 6, include: flange thickness: d, d 1 、d 2 、d 3 、d 4 The method comprises the steps of carrying out a first treatment on the surface of the Flange height: l (L) 1 、l 2 、l 3 、l 4 The method comprises the steps of carrying out a first treatment on the surface of the Height of H-section steel: d, d 5 And d 6 The method comprises the steps of carrying out a first treatment on the surface of the H-section web thickness: d, d 9 And d 10 The method comprises the steps of carrying out a first treatment on the surface of the Width of H-shaped steel: d, d 7 And d 8 。
According to the measurement scheme, four independent cameras and laser models are adopted to measure the cross section size of the H-shaped steel, the cross section size is not interfered with each other during working, errors caused by cooperative work are avoided, the overall measurement accuracy is high, and the extraction accuracy of the light bar center line aiming at the irregular light bar shape on the surface of the H-shaped steel is optimized based on a local regression analysis method and a target pixel set forming the light bar center line in a secondary fitting corrected extracted light bar image; the local regression analysis only needs to frame and select the field with the preset field radius around the pixel points of the rough extracted light bar center line, then regression analysis is carried out, the operation amount is reduced, the extraction speed of a target pixel point set of the light bar center line is improved, the light bar cross section direction corresponding to the local pixel points of each light bar center line is worked out through a linear regression formula, then the pixel points in the direction are processed, the extraction precision of the light bar center line aiming at the light bar shape on the surface of the H-shaped steel is further optimized, the light bar center line is smoother, the extracted target pixel point set of the light bar center line is more accurate, and the measurement precision of the cross section size of the H-shaped steel is further improved; and meanwhile, the normal line and the tangential direction of the pixel point of the central line of each light bar can be output, and some characteristic points in the section size measurement of the H-shaped steel can be extracted by taking the normal line and the tangential direction as constraint conditions.
The first direction light bar image and the second direction light bar image correspond to the same dimension measuring method, the third direction light bar image and the fourth direction light bar image correspond to the same dimension measuring method, and the calculation of the first direction light bar image and the fourth light bar image is taken as an example for specific explanation.
As an alternative way, due to uneven surface of the H-shaped steel, the broken line condition of the light bar in the photographed light bar image may occur, as shown in fig. 7, and the broken line condition occurs at the right side inclined angle of the light bar at room temperature, as shown in fig. 8, and the broken line condition of the light bar is more obvious at high temperature. The light bar image is processed, and the ROI area image is extracted from the light bar image in order to reduce the time of the subsequent image processing, specifically, the light bar image is firstly subjected to expansion processing, so that the light bars in the light bar image are connected into a whole; when the broken light line is stitched, whether adjacent interference points exist in the image is observed, if so, the interference points are corroded, and an appropriate expansion template can be selected for expansion treatment according to the broken light line degree.
Screening out the area with the largest area containing the light bar by adopting an area screening method, and calculating the minimum circumscribed rectangle of the area with the largest area to obtain an ROI area image: as shown in fig. 9, the image within the rectangular frame is the extracted ROI area image; the ROI area image is used for extracting a target pixel point set of the light bar center line and determining a first characteristic point and a second characteristic point.
After the ROI area where the light bar is located is determined, the target pixel point set of the light bar center line is extracted by applying the light bar center line extraction algorithm based on the partial regression analysis and the quadratic fitting correction to the area. The extraction algorithm is described in conjunction with fig. 10, and specifically includes the following steps:
preprocessing light bars, and filtering and denoising the image: carrying out Gaussian convolution on the ROI region image, removing image noise points, and processing the light bar into a Gaussian model;
pixel level coarse extraction of pixel points of the light bar center line: defining a gray threshold value, traversing all pixel points of the ROI area image, and determining the pixel points larger than the gray threshold value as the pixel points of the upper and lower boundaries of the light bars in the light bar image corresponding to each direction; determining a first pixel point set of a light bar central line according to the pixel points of the upper and lower boundaries, wherein the geometric center of the upper and lower boundaries is the approximate position of the light bar central line, and the coordinates of the first pixel point set of the light bar central line which is extracted in a rough way are set as (t) x ,t y );
Correcting the first pixel point set based on a local regression analysis method and a linear regression formula to obtain the normal direction corresponding to each pixel point in the first pixel point set: firstly, traversing each pixel point in a first pixel point set of a rough extracted light bar central line, and determining an a×a neighborhood corresponding to each pixel point in the first pixel point set by taking the pixel point of the rough extracted light bar central line as a center according to a preset neighborhood radius, wherein the neighborhood is a a×a field selected by taking one pixel point as a center as shown in fig. 11; then defining a new gray threshold, screening out pixel points meeting preset pixel conditions from each pixel point neighborhood in the first pixel point set, setting the pixel point larger than the gray threshold as 1, and setting the pixel point smaller than the gray threshold as 0: specifically, as shown in fig. 12, a pixel with a gray value greater than the gray threshold is set to 1, a pixel with a gray value less than the gray threshold is set to 0, a pixel with a 1 marked as a pixel satisfying a preset pixel condition is set as a pixel, a coordinate with a 1 element is found out for regression analysis, and the trend of the pixel in the field is fitted according to a linear regression formula, namely the tangential direction The linear regression formula is shown as formula1) The following is shown:
(1)
wherein,is the tangential direction of the local pixel point, +.>The number of the pixel points is 1, and the gray value of the pixel points is +.>Pixel sequence number of 1 for the current element, < >>Is the transverse coordinate of the pixel point, which is->Is the longitudinal coordinate of the pixel point, < >>Is the average value of the transverse coordinates of the pixel points, +.>Is the average value of the longitudinal coordinates of the pixel points.
The normal direction of the point. The normal direction of part of the pixels of the first pixel point set of the center line of the light bar extracted in a rough way is shown in fig. 13, and the arrow pointing is the normal direction of the pixels.
Calculating offset by sub-pixel level parabolic fitting, and performing parabolic fitting along the main direction of the normal to obtain the offset: determining the offset of the first pixel point set based on a parabolic fitting method; specifically, after determining the local normal direction, continuing to find all pixel points through which the normal passes in the field, extracting all pixel points through which the normal passes in the field of each pixel point in the first pixel point set, wherein the extracted gray values of the pixel points pass the field and the threshold limit, so that the light bar section data formed by the extracted pixel points are approximate to a parabola, and fitting all pixel points through which the normal passes in the field into a parabola along the normal direction, and determining the coordinate value of the center point of the parabola as an offset; referring to fig. 14, the abscissa value of the parabolic center point is the value of the determined offset. Determining a target pixel point set of the light bar center line according to the offset and the first pixel point set of the light bar center line which is extracted in a rough mode;
In order to further improve the accuracy of the target pixel point set of the extracted light bar center line, a gray level gravity center method can be adopted to determine the sub-offset, the offset is marked as a main offset, specifically, gray values of all pixel points through which the normal passes are extracted to obtain a gray level value set, and intermediate values in the gray level value set are used as the sub-offset. The two offset amounts can be described with reference to fig. 14, as shown in fig. 14, the parabola is a parabola fitted by all the pixel points passing through the normal line, the abscissa is the coordinate value of the pixel point in the normal line direction, the ordinate is the gray value of the pixel point, and the x-direction coordinate value of the central point of the parabola in the horizontal direction is taken as the main offset amountThe method comprises the steps of carrying out a first treatment on the surface of the The gray value of the middle point in the gray value set of each parabolic point in the ordinate direction is the partial offset. The coordinate offset of the center line of the rough extracted light bar relative to the center line of the original light bar is as follows。
Solving the accurate coordinates of the pixel points of the center line of the light bar: and determining a second pixel point set forming the central line of the light bar according to the first pixel point set and the offset, and determining the second pixel point set as a target pixel point set of the central line of the light bar. Specifically, whether the offset is larger than the neighborhood size is determined, i.e., the first offset component is determined And a second offset componentIf the offset is larger than the radius of the preset field, setting the offset to zero, and if the offset is smaller than the radius of the preset field, determining the accurate position of the target pixel set of the light bar center line according to the pixel set of the rough extraction light bar center line and the offset, wherein the coordinate is as follows. The main offset can correct the pixel point set of the center line of the rough extracted light bar in the vertical direction, so that the center line of the obtained light bar is smoother.
After the target pixel point set of the center line of the light bar is extracted, combining the requirement of the H-shaped steel on the dimension to be measured, starting to extract key feature points of the first-direction light bar image and the second-direction light bar image; first, first feature points of a first-direction light bar image are extracted: the width of the H-shaped steel is calculated at a first boundary point of the flange, namely a lower left edge point and a lower right edge point;
taking a first-direction light bar image as an example for explanation, specifically, denoising an ROI (region of interest) area image of the light bar image, setting an image threshold to be 0.5-0.9, performing binarization, screening out noise points in the ROI area image by an area screening method, and removing the noise points to obtain the ROI area image corresponding to the denoised light bar image;
traversing all pixel points in the ROI area image corresponding to the light bar image, and determining a flange first boundary point of the H-shaped steel; specifically, finding out the pixel points with gray values of 255 at the leftmost end and the rightmost end in the ROI area, and marking the pixel points as points c1 and c 2; selecting a proper rectangular template according to the width of the light bar, taking c1 and c2 as centers in the range of the rectangular template, finding out the pixel points with the gray values of 255 in the neighborhood of c1 and c2, and selecting the pixel points with the gray values of 255 at the lowest and the leftmost in the neighborhood; and meanwhile, the condition that the number of candidate points screened out from the lowest part and the leftmost part is greater than or equal to two is satisfied, the intersection point of a row formed by the pixel points selected from the lowest part and a column formed by the pixel points selected from the leftmost part is marked as a point 1 and a point 2, the point 1 and the point 2 are first characteristic points extracted from the first direction light bar image, and the flange is a first boundary point.
And then extracting second characteristic points of the first-direction light bar image: a flange second boundary point, a flange center point and four characteristic points of the web plate;
specifically, the ROI area image of the light bar image is subjected to block processing, the ROI area image is equally divided into four areas through the center point of the ROI area image, and the four areas are marked as I, II, III, IV areas in turn anticlockwise;
according to the extracted target pixel point set of the light bar center point and the corresponding slope of each pixel point in the target pixel point set, selecting a proper slope threshold value according to the inclination angle of the light bar, and screening out three sections of horizontal light stripes of the light bar in the light bar image; the three-section horizontal light stripe comprises a first section horizontal light stripe, a second section horizontal light stripe and a third section horizontal light stripe, and the lengths of the first section horizontal light stripe and the second section horizontal light stripe are smaller than those of the third section horizontal light stripe; the first section of horizontal light stripe is in the area I, the second section of horizontal light stripe is in the area IV, and the third section of horizontal light stripe spans the area II and the area III.
Screening the middle lattice of the first section of horizontal light stripes and the second section of horizontal light stripes in the I area and the IV area in the horizontal direction and the vertical direction respectively, removing points with larger errors, and fitting in the flange thickness direction of the H-shaped steel by adopting a least square method to generate a first straight line and a second straight line; the straight line effect is shown in fig. 15, wherein the straight line in the figure is a second straight line fitted by a second section of horizontal light stripe, and the light stripe in the horizontal direction in the figure is an H-shaped steel flange.
Idealizing the H-shaped steel flange thickness models in the area I and the area IV into a roof model, wherein the roof model is shown in fig. 16, and the two sides of the roof model are symmetrical arcs;
determining a flange second boundary point and a flange center point according to the first straight line, the second straight line and the flange first boundary point: and translating the first straight line to pass point 1, translating the second straight line to pass point 2, and obtaining points with gray values of 255 at the leftmost end in the I area and the rightmost end in the IV area along the straight line, namely points 5 and 6, wherein the points 5 and 6 are flange second boundary points. The center point of the line segment formed by the midpoint 1 and the point 5 of the first straight line is selected as a point 3, the center point of the line segment formed by the midpoint 2 and the point 6 of the second straight line is selected as a point 4, and the center points of the flanges are selected as the center points of the flanges.
Selecting the center point of a third section of horizontal light stripe in the II and III regions, deleting the middle line segment of the third section of horizontal light stripe, screening error points of the remaining two line segments through a median lattice, and performing straight line fitting on the remaining line segments to obtain a third straight line and a fourth straight line; the third section of horizontal light stripe is divided into three sections, and the length of the middle line section is one half of the length of the third section of horizontal light stripe; the screening of the median set is determined according to the extraction effect of the target pixel point set of the light bar center line and the number of the extraction points.
Four characteristic points of the web are determined according to the third straight line and the fourth straight line: 4 points in the fitted third line and fourth line are selected and marked as point 7, point 8, point 9 and point 10, and two points are selected on each line.
And extracting the first characteristic points and the second characteristic points of the second-direction light bar image by adopting the same method.
Finally, in combination with the requirement of the H-shaped steel on the dimension to be measured, extracting the first characteristic points of the third light bar image and the fourth light bar image, wherein the extraction method of the first characteristic points of the third light bar image and the fourth light bar image is the same as that of the first characteristic points of the first light bar image:
taking a fourth-direction light bar image as an example for explanation, specifically, denoising an ROI (region of interest) area image of the light bar image, setting an image threshold to be 0.5-0.9, performing binarization, screening out noise points in the ROI area image by an area screening method, and removing the noise points to obtain the ROI area image corresponding to the denoised light bar image;
traversing all pixel points in the ROI area image corresponding to the light bar image, and determining edge points of the side face of the H-shaped steel: finding out the pixel points with gray values of 255 at the leftmost end and the rightmost end in the ROI area, and marking the pixel points as c11 and c12 points; selecting a proper rectangular template according to the width of the light bar, taking c11 and c12 as centers in the range of the rectangular template, finding out the pixel points with the gray values of 255 in the neighborhood of c11 and c12, and selecting the pixel points with the values of 255 at the bottommost and leftmost Fang Huidu in the neighborhood; and meanwhile, the condition that the number of candidate points screened out from the lowest part and the leftmost part is more than or equal to two is satisfied, the intersection point of a row formed by the pixel points selected from the lowest part and a column formed by the pixel points selected from the leftmost part is marked as a point 11 and a point 12, the point 11 and the point 12 are the first characteristic points extracted from the fourth-direction light bar image, and the edge points of the side face of the H-shaped steel. And determining a first characteristic point of the third-direction light bar image by adopting the same method of the fourth-direction light bar image, and determining edge points of the side surface of the H-shaped steel.
After the key feature points are extracted, combining camera external parameters obtained after calibration, and performing space transformation on the first feature points and the second feature points to obtain actual coordinates; for the conventional size without a vision blind area, measuring the width of the H-shaped steel, the thickness of the flange and the height of the flange by using an upper camera and a lower camera; the left and right cameras measure the height dimension of the H-beam. For the web size in the visual blind area, according to the shape characteristics of the H-shaped steel, the flange height measured by the upper camera and the lower camera is subtracted from the H-shaped steel height measured by the left camera and the right camera to indirectly measure the thickness of the H-shaped steel web.
Firstly, determining the flange thickness, the H-shaped steel width and the H-shaped steel height by adopting a distance calculation formula (2) between two points based on the flange first boundary point, the flange second boundary point and the edge point of the side surface of the H-shaped steel; specifically, the thickness dimension d of the flange of the H-shaped steel is calculated according to the point 1 and the point 5 1 Obtaining the thickness dimension d of the flange of the H-shaped steel according to the point 2 and the point 6 2 The method comprises the steps of carrying out a first treatment on the surface of the H-shaped steel height dimension d is calculated according to point 11 and point 12 6 The method comprises the steps of carrying out a first treatment on the surface of the H-type width dimension d is calculated according to point 1 and point 2 7 。
Distance between two pointsThe calculation formula of (2) is shown as the formula:
(2)
wherein, the method comprises the following steps of,) And%,) Is the coordinates of the extracted feature points.
Determining the flange height of the H-shaped steel by adopting a calculation formula (3) of the distance from the point to the straight line based on a straight line vector formed by the flange center point and four characteristic points of the web plate; if the height dimension l of the flange of the H-shaped steel is determined according to the straight line vector formed by the points 3 and 4 and the points 7, 9, 8 and 10 1 And l 2 ;
The calculation formula of the distance from the point to the straight line is shown in formula (3):
(3)
wherein,as the distance from the point to the straight line,for a vector consisting of a point outside the line and a starting point on the line,a straight line vector determined for both the start point and the end point.
Determining the thickness of an H-shaped steel web plate according to the height dimension of the H-shaped steel and the height dimension of the flange;
the thickness of the web plate of the H-shaped steel is finished by adopting a formula (4) according to the shape characteristics of the H-shaped steelAndmeasurement of l 1 、l 2 、l 3 、l 4 Is the flange height, d 5 And d 6 For H-section steel height, equation (4) is as follows:
(4)
and (5) finishing the measurement of the section size of the H-shaped steel.
Taking an example of H-shaped steel with the specification of 200 x 100 as an example, the measurement results are shown in table 1:
table 1 section size measurement results of H section steel
By the description of the H-shaped steel dimension measuring method, the invention has the following effects:
the invention adopts four independent camera-laser models to measure the dimension of the H-shaped steel, the working is not interfered with each other, the error caused by cooperative work is not generated, and the whole measuring precision is high.
The light bar center extraction algorithm based on the partial regression analysis and the quadratic fitting correction optimizes the light bar center extraction precision aiming at the irregular light bar shape of the H-shaped steel surface; meanwhile, the normal line and the tangential direction of each center point can be output, and some characteristic points in the measurement of the H-shaped steel size can be extracted by taking the normal line and the tangential direction as constraint conditions.
According to the invention, the ROI is extracted by adopting an area screening method according to the image characteristics, the extraction speed is high, the image processing time can be greatly reduced, and the real-time property of measurement is improved; meanwhile, compared with the traditional center extraction method considering the light bar direction, the light bar center extraction algorithm based on the local regression analysis and the quadratic fitting correction improves the extraction speed by nearly 4 times; through actual measurement, the average processing time of each image is 0.25 seconds, and the real-time requirement of H-shaped steel measurement is met.
The key feature extraction of the invention adopts three sections of horizontal light stripes to calculate the key dimensions above and below according to the shape features of the H-shaped steel, thereby avoiding the situation that the light stripes cannot be measured normally due to broken lines.
The key feature extraction of the invention extracts according to the shape features of the H-shaped steel, solves the defect of shape template matching, and is applicable to the measurement of the sizes of H-shaped steel with various specifications.
The key feature extraction method eliminates the point with larger error by screening the median point set, and fits and calculates the point in a segmented way, thereby improving the measurement precision and robustness.
The invention solves the problems of low efficiency and low measurement precision of the traditional manual measurement and realizes the high-precision real-time measurement of the H-shaped steel dimension.
The present invention may actually perform the division of the functional modules according to the above-described method example, for example, each functional module may be divided corresponding to each function, or two or more functions may be integrated in one processing unit. The integrated modules may be implemented in hardware or in software functional modules. It should be noted that, in the embodiment of the present invention, the division of the modules is schematic, which is merely a logic function division, and other division manners may be implemented in actual implementation.
Fig. 17 shows a schematic structural diagram of an H-section steel dimension measuring apparatus provided by the present invention in the case of dividing each functional module by corresponding each function. As shown in fig. 17, the apparatus may include:
a light bar image acquisition module 171 for the H-section steel for acquiring a light bar image of the H-section steel; the light bar image is an image formed by irradiating the H-shaped steel with a laser; the light bar images comprise a first direction light bar image, a second direction light bar image, a third direction light bar image and a fourth direction light bar image;
The target pixel point set determining module 172 is configured to extract a target pixel point set that forms a light bar center line in the light bar image based on a local regression analysis method and a quadratic fit correction;
a first feature point determining module 173, configured to extract first feature points that satisfy a preset condition in the light bar image corresponding to each direction;
a second feature point determining module 174, configured to determine a second feature point in the first direction light bar image and the second direction light bar image based on the first feature point, the target pixel point set, and the slopes corresponding to the target pixel point set;
and the section size calculation module 175 of the H-shaped steel is used for determining the section size of the H-shaped steel according to the first characteristic points and the second characteristic points.
Optionally, the H-section steel comprises a flange and a web; the first feature point includes: a flange first boundary point corresponding to the first direction light bar image and the second direction light bar image, and an edge point of the side surface of the H-shaped steel corresponding to the third direction light bar image and the fourth direction light bar image; the second feature point includes: the flange center point and the flange second boundary point corresponding to the first direction light bar image and the second direction light bar image and the four characteristic points of the web plate; the second feature point determining module 174 may include:
The three-section horizontal light stripe screening unit is used for screening out three sections of horizontal light stripes of the light stripe in the light stripe image according to the target pixel point set and the slope corresponding to the target pixel point set; the three-section horizontal light stripe comprises a first section horizontal light stripe, a second section horizontal light stripe and a third section horizontal light stripe, and the lengths of the first section horizontal light stripe and the second section horizontal light stripe are smaller than those of the third section horizontal light stripe;
the first straight line and second straight line fitting unit is used for fitting the first section of horizontal light stripes and the second section of horizontal light stripes in the flange thickness direction of the H-shaped steel by adopting a least square method to generate a first straight line and a second straight line;
a flange second boundary point and flange center point determining unit configured to determine a flange second boundary point and a flange center point according to the first straight line, the second straight line, and the flange first boundary point;
the third straight line and fourth straight line fitting unit is used for deleting the middle line segment of the third section of horizontal light stripe, and performing straight line fitting on the rest line segments to obtain a third straight line and a fourth straight line; the third section of horizontal light stripe is divided into three sections, and the length of the middle line section is one half of the length of the third section of horizontal light stripe;
And the four characteristic point determining units are used for determining four characteristic points of the web according to the third straight line and the fourth straight line.
Optionally, the first feature point determining module 173 may include:
the denoising processing unit is used for denoising the light bar image corresponding to each direction to obtain a denoised light bar image;
the flange first boundary point determining unit is used for traversing the pixel points in the light bar image and determining the flange first boundary point of the H-shaped steel;
and the flange second boundary point determining unit is used for traversing pixel points in the light bar image and determining edge points of the side surface of the H-shaped steel.
Optionally, the target pixel set determining module 172 may include:
a determining unit for determining the pixel points of the upper and lower boundaries of the light bar in the light bar image according to the gray threshold;
the first pixel point set determining unit is used for determining a first pixel point set of the central line of the light bar according to the pixel points of the upper boundary and the lower boundary;
the correction unit is used for correcting the first pixel point set based on a local regression analysis method and a linear regression formula to obtain the normal direction corresponding to each pixel point in the first pixel point set;
And the offset determining unit is used for determining the offset of the first pixel point set by adopting a parabolic fitting method according to the normal direction corresponding to each pixel point in the first pixel point set.
Optionally, the correction unit may specifically be configured to:
determining the neighborhood of each pixel point in the first pixel point set according to a preset neighborhood radius;
selecting pixel points meeting preset pixel conditions from the neighborhood;
fitting the trend of the pixel points meeting the preset pixel condition in the neighborhood according to a linear regression formula to obtain the normal direction corresponding to each pixel point in the first pixel point set.
Optionally, the offset determining unit may specifically be configured to:
extracting all pixel points through which normal lines in the neighborhood of each pixel point in the first pixel point set pass;
and fitting all pixel points through which the normal line passes into a parabola in the normal line direction, and determining the value of the central point of the parabola as an offset.
Optionally, the section size calculating module 175 of the H-section steel may include:
the flange thickness, the width of the H-shaped steel and the height of the H-shaped steel are determined by adopting a distance calculation formula between the two points based on the flange first boundary point, the flange second boundary point and the edge point of the side face of the H-shaped steel;
The flange height determining unit is used for determining the flange height of the H-shaped steel by adopting a calculation formula of the distance from the point to the straight line based on a straight line vector formed by the flange center point and four characteristic points of the web;
and the H-shaped steel web thickness determining unit is used for determining the thickness of the H-shaped steel web according to the height of the H-shaped steel and the height of the flange.
Optionally, the apparatus further includes a light bar image processing module, and the light bar image processing module may include:
the expansion processing unit is used for carrying out expansion processing on the light bar image so that light bars in the light bar image are connected into a whole;
the area screening unit is used for screening out the area with the largest area containing the light bar by adopting an area screening method, calculating the minimum circumscribed rectangle of the area with the largest area, and obtaining an ROI area image, wherein the ROI area image is used for extracting the target pixel point set of the central line of the light bar and determining the first characteristic point and the second characteristic point.
Optionally, the light bar image acquisition module 171 of the H-section steel may include:
the laser and camera placing unit is used for placing the laser and the camera in four preset directions of the H-shaped steel respectively; the laser emitted by the laser vertically irradiates the surface of the H-shaped steel, and the included angle between the camera and the laser is a preset included angle;
The calibration unit is used for calibrating the camera by adopting a Zhang Zhengyou calibration method to obtain a camera external parameter; the camera external parameters are used for converting the first characteristic points and the second characteristic points into world coordinates.
The above description has been presented mainly in terms of interaction between the modules, and the solution provided by the embodiment of the present invention is described. It is to be understood that, in order to achieve the above-described functions, they comprise corresponding hardware structures and/or software modules that perform the respective functions. Those of skill in the art will readily appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as hardware or combinations of hardware and computer software. Whether a function is implemented as hardware or computer software driven hardware depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
In the above embodiments, it may be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present invention are performed in whole or in part. The computer may be a general purpose computer, a special purpose computer, a computer network, a terminal, a user equipment, or other programmable apparatus. The computer program or instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another computer readable storage medium, for example, the computer program or instructions may be transmitted from one website site, computer, server, or data center to another website site, computer, server, or data center by wired or wireless means. The computer readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server, data center, etc. that integrates one or more available media. The usable medium may be a magnetic medium, e.g., floppy disk, hard disk, tape; optical media, such as digital video discs (digital video disc, DVD); but also semiconductor media such as solid state disks (solid state drive, SSD).
Although the invention is described herein in connection with various embodiments, other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the claims, the word "comprising" does not exclude other elements or steps, and the "a" or "an" does not exclude a plurality. A single processor or other unit may fulfill the functions of several items recited in the claims. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.
Although the invention has been described in connection with specific features and embodiments thereof, it will be apparent that various modifications and combinations can be made without departing from the spirit and scope of the invention. Accordingly, the specification and drawings are merely exemplary illustrations of the present invention as defined in the appended claims and are considered to cover any and all modifications, variations, combinations, or equivalents that fall within the scope of the invention. It will be apparent to those skilled in the art that various modifications and variations can be made to the present invention without departing from the spirit or scope of the invention. Thus, it is intended that the present invention also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (10)
1. A method for measuring the dimensions of H-steel, comprising:
acquiring a light bar image of the H-shaped steel; the light bar image is an image formed by irradiating the H-shaped steel with a laser; the light bar images comprise a first direction light bar image, a second direction light bar image, a third direction light bar image and a fourth direction light bar image;
extracting a target pixel point set forming a light bar center line in the light bar image based on a local regression analysis method and secondary fitting correction;
extracting first characteristic points meeting preset conditions in the light bar image corresponding to each direction;
determining second characteristic points in the first direction light bar image and the second direction light bar image based on the first characteristic points, the target pixel point set and slopes corresponding to the target pixel point set;
and determining the section size of the H-shaped steel according to the first characteristic points and the second characteristic points.
2. The H-section steel dimension measuring method according to claim 1, wherein the H-section steel includes a flange and a web; the first feature point includes: a flange first boundary point corresponding to the first direction light bar image and the second direction light bar image, and an edge point of the side surface of the H-shaped steel corresponding to the third direction light bar image and the fourth direction light bar image; the second feature point includes: the flange center point and the flange second boundary point corresponding to the first direction light bar image and the second direction light bar image and the four characteristic points of the web plate; the determining the second feature points of the first direction light bar image and the second direction light bar image based on the first feature points, the target pixel point set and the slopes corresponding to the target pixel point set comprises:
Screening three sections of horizontal light stripes of the light stripe in the light stripe image according to the slope corresponding to the target pixel point set; the three-section horizontal light stripe comprises a first section horizontal light stripe, a second section horizontal light stripe and a third section horizontal light stripe, and the lengths of the first section horizontal light stripe and the second section horizontal light stripe are smaller than those of the third section horizontal light stripe;
fitting the first section of horizontal light stripes and the second section of horizontal light stripes in the thickness direction of the flange of the H-shaped steel by adopting a least square method to generate a first straight line and a second straight line;
determining a flange second boundary point and a flange center point according to the first straight line, the second straight line and the flange first boundary point;
deleting the middle line segment of the third section of horizontal light stripe, and performing straight line fitting on the rest line segments to obtain a third straight line and a fourth straight line; the third section of horizontal light stripe is divided into three sections, and the length of the middle line section is one half of the length of the third section of horizontal light stripe;
and determining four characteristic points of the web plate according to the third straight line and the fourth straight line.
3. The method for measuring the size of the H-beam according to claim 2, wherein extracting the first feature point meeting the preset condition in the light bar image corresponding to each direction comprises:
Denoising the light bar image corresponding to each direction to obtain a denoised light bar image;
traversing pixel points in the light bar image, and determining a flange first boundary point of the H-shaped steel;
and traversing pixel points in the light bar image, and determining edge points of the side surface of the H-shaped steel.
4. The H-beam dimension measurement method according to claim 1, wherein the extracting the target pixel point set constituting the light bar center line in the light bar image based on the partial regression analysis and the quadratic fit correction comprises:
determining pixel points of upper and lower boundaries of the light bars in the light bar image according to the gray threshold;
determining a first pixel point set of a light bar central line according to the pixel points of the upper and lower boundaries;
correcting the first pixel point set based on a local regression analysis method and a linear regression formula to obtain a normal direction corresponding to each pixel point in the first pixel point set;
determining the offset of the first pixel point set by adopting a parabolic fitting method according to the normal direction corresponding to each pixel point in the first pixel point set;
and determining a second pixel point set forming the central line of the light bar according to the first pixel point set and the offset, and determining the second pixel point set as a target pixel point set of the central line of the light bar.
5. The H-beam dimension measurement method according to claim 4, wherein the correcting the first pixel point set based on the local regression analysis method and the linear regression formula to obtain the normal direction corresponding to each pixel point in the first pixel point set comprises:
determining the neighborhood of each pixel point in the first pixel point set according to a preset neighborhood radius;
selecting pixel points meeting preset pixel conditions from the neighborhood;
fitting the trend of the pixel points meeting the preset pixel condition in the neighborhood according to a linear regression formula to obtain the normal direction corresponding to each pixel point in the first pixel point set.
6. The method for measuring the size of the H-beam according to claim 5, wherein determining the offset of the first pixel set by using a parabolic fitting method according to the normal direction corresponding to each pixel in the first pixel set comprises:
extracting all pixel points through which normal lines in the neighborhood of each pixel point in the first pixel point set pass;
and fitting all pixel points through which the normal line passes into a parabola in the normal line direction, and determining the value of the central point of the parabola as an offset.
7. The H-section steel dimension measuring method according to claim 2, wherein the determining the section dimension of the H-section steel from the first characteristic point and the second characteristic point comprises:
Determining the flange thickness, the H-shaped steel width and the H-shaped steel height by adopting a distance calculation formula between the two points based on the flange first boundary point, the flange second boundary point and the edge point of the side surface of the H-shaped steel;
determining the flange height of the H-shaped steel by adopting a calculation formula of the distance from the point to the straight line based on a straight line vector formed by the flange center point and four characteristic points of the web plate;
and determining the thickness of the H-shaped steel web plate according to the height of the H-shaped steel and the height of the flange, and finishing the determination of the section size of the H-shaped steel.
8. The H-section steel dimension measuring method according to claim 1, wherein the acquiring the light bar image of the H-section steel further comprises:
expanding the light bar image to enable light bars in the light bar image to be connected into a whole;
and screening out a region with the largest area containing the light bar by adopting an area screening method, and calculating the minimum circumscribed rectangle of the region with the largest area to obtain an ROI region image, wherein the ROI region image is used for extracting a target pixel point set of the central line of the light bar and determining a first characteristic point and a second characteristic point.
9. The H-section steel dimension measuring method according to claim 1, wherein the acquiring the light bar image of the H-section steel comprises:
Respectively placing a laser and a camera in four preset directions of the H-shaped steel; the laser emitted by the laser vertically irradiates the surface of the H-shaped steel, and the included angle between the camera and the laser is a preset included angle;
calibrating the camera by adopting a Zhang Zhengyou calibration method to obtain a camera external parameter; the camera external parameters are used for converting the first characteristic points and the second characteristic points into world coordinates.
10. An H-section steel dimension measuring device, comprising:
the light bar image acquisition module of the H-shaped steel is used for acquiring light bar images of the H-shaped steel; the light bar image is an image formed by irradiating the H-shaped steel with a laser; the light bar images comprise a first direction light bar image, a second direction light bar image, a third direction light bar image and a fourth direction light bar image;
the target pixel point set determining module is used for extracting a target pixel point set forming a light bar center line in the light bar image based on a local regression analysis method and secondary fitting correction;
the first characteristic point determining module is used for extracting first characteristic points meeting preset conditions in the light bar image corresponding to each direction;
the second characteristic point determining module is used for determining second characteristic points in the first direction light bar image and the second direction light bar image based on the first characteristic points, the target pixel point set and slopes corresponding to the target pixel point set;
And the section size calculation module is used for determining the section size of the H-shaped steel according to the first characteristic points and the second characteristic points.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311516536.4A CN117237434B (en) | 2023-11-15 | 2023-11-15 | H-shaped steel size measurement method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202311516536.4A CN117237434B (en) | 2023-11-15 | 2023-11-15 | H-shaped steel size measurement method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN117237434A true CN117237434A (en) | 2023-12-15 |
CN117237434B CN117237434B (en) | 2024-02-09 |
Family
ID=89084726
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202311516536.4A Active CN117237434B (en) | 2023-11-15 | 2023-11-15 | H-shaped steel size measurement method and device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN117237434B (en) |
Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107657377A (en) * | 2017-09-26 | 2018-02-02 | 大连理工大学 | A kind of public transportation lane policy evaluation method returned based on breakpoint |
CN108489383A (en) * | 2018-03-01 | 2018-09-04 | 北京科技大学 | A kind of measuring device and method of H-type cross dimensions |
CN108765419A (en) * | 2018-06-20 | 2018-11-06 | 天津大学 | Structure light vision weld image information self-adapting extracting method |
CN108921864A (en) * | 2018-06-22 | 2018-11-30 | 广东工业大学 | A kind of Light stripes center extraction method and device |
CN109145354A (en) * | 2017-06-18 | 2019-01-04 | 科文托尔公司 | The system and method for key parameter identification, process modeling calibration and Mutability analysis |
CN109272544A (en) * | 2018-10-22 | 2019-01-25 | 湖北文理学院 | The structural light three-dimensional measurement model and image processing method of pipeline all-position weld seam |
WO2022116218A1 (en) * | 2020-12-05 | 2022-06-09 | 浙江大学 | Method for extracting line-structured laser center for complex surface |
WO2022143796A1 (en) * | 2020-12-29 | 2022-07-07 | 杭州海康机器人技术有限公司 | Calibration method and calibration device for line structured light measurement system, and system |
US20230084728A1 (en) * | 2020-06-19 | 2023-03-16 | Zhejiang Huaray Technology Co., Ltd. | Systems and methods for object measurement |
CN116105607A (en) * | 2023-02-03 | 2023-05-12 | 太原理工大学 | Full-size measuring device and method for non-contact cold-hot H-shaped steel |
-
2023
- 2023-11-15 CN CN202311516536.4A patent/CN117237434B/en active Active
Patent Citations (10)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109145354A (en) * | 2017-06-18 | 2019-01-04 | 科文托尔公司 | The system and method for key parameter identification, process modeling calibration and Mutability analysis |
CN107657377A (en) * | 2017-09-26 | 2018-02-02 | 大连理工大学 | A kind of public transportation lane policy evaluation method returned based on breakpoint |
CN108489383A (en) * | 2018-03-01 | 2018-09-04 | 北京科技大学 | A kind of measuring device and method of H-type cross dimensions |
CN108765419A (en) * | 2018-06-20 | 2018-11-06 | 天津大学 | Structure light vision weld image information self-adapting extracting method |
CN108921864A (en) * | 2018-06-22 | 2018-11-30 | 广东工业大学 | A kind of Light stripes center extraction method and device |
CN109272544A (en) * | 2018-10-22 | 2019-01-25 | 湖北文理学院 | The structural light three-dimensional measurement model and image processing method of pipeline all-position weld seam |
US20230084728A1 (en) * | 2020-06-19 | 2023-03-16 | Zhejiang Huaray Technology Co., Ltd. | Systems and methods for object measurement |
WO2022116218A1 (en) * | 2020-12-05 | 2022-06-09 | 浙江大学 | Method for extracting line-structured laser center for complex surface |
WO2022143796A1 (en) * | 2020-12-29 | 2022-07-07 | 杭州海康机器人技术有限公司 | Calibration method and calibration device for line structured light measurement system, and system |
CN116105607A (en) * | 2023-02-03 | 2023-05-12 | 太原理工大学 | Full-size measuring device and method for non-contact cold-hot H-shaped steel |
Non-Patent Citations (3)
Title |
---|
KU CHIN LIN 等: "Measurement of the cross-sectional contour of H-shaped steel using multiple stereo pairs", 《2005 ICSC CONGRESS ON COMPUTATIONAL INTELLIGENCE METHODS AND APPLICATIONS》, pages 1 - 6 * |
刘庆纲: "H型钢端面尺寸的精密视觉检测方法", 《光电工程》, vol. 40, no. 11, pages 1 - 7 * |
刘燕: "H型钢截面尺寸线结构光测量", 《中国优秀硕士学位论文全文数据库信息科技辑》, pages 140 - 447 * |
Also Published As
Publication number | Publication date |
---|---|
CN117237434B (en) | 2024-02-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112651968B (en) | Wood board deformation and pit detection method based on depth information | |
JP5421192B2 (en) | Crack detection method | |
JP6099479B2 (en) | Crack detection method | |
CN109583365B (en) | Method for detecting lane line fitting based on imaging model constrained non-uniform B-spline curve | |
JP5852919B2 (en) | Crack detection method | |
JP5303405B2 (en) | Vehicle inspection device | |
JP7008409B2 (en) | Crack detection method | |
JP6811217B2 (en) | Crack identification method, crack identification device, crack identification system and program on concrete surface | |
JP2009199575A (en) | Image matching method, program and applied device | |
CN109829856B (en) | Bridge crack information fusion method | |
CN111145157A (en) | Road network data automatic quality inspection method based on high-resolution remote sensing image | |
JP5812705B2 (en) | Crack detection method | |
JP5705711B2 (en) | Crack detection method | |
JP2018128309A (en) | Crack detection method | |
CN114972575A (en) | Linear fitting algorithm based on contour edge | |
CN114612412A (en) | Processing method of three-dimensional point cloud data, application of processing method, electronic device and storage medium | |
KR101492505B1 (en) | Method for Evaluating the State Information of Rebar for the Reinforced Concrete using the Photo Image of Rebar Assembly | |
CN116188544A (en) | Point cloud registration method combining edge features | |
CN117649451B (en) | Visual positioning method for positioning edge finder wafer | |
CN107657620A (en) | A kind of method and system of textured metal freezing region recognition | |
CN117237434B (en) | H-shaped steel size measurement method and device | |
CN112184784B (en) | Dual-spectrum image alignment method and device, electronic equipment and storage medium | |
CN111476792B (en) | Extraction method of strip steel image contour | |
CN114842041A (en) | Embedded-based steel rail profile detection light bar extraction method | |
CN115018735A (en) | Fracture width identification method and system for correcting two-dimensional code image based on Hough transform |
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 |