CN113888446A - Intelligent detection method for bending line of sheet metal structural part - Google Patents
Intelligent detection method for bending line of sheet metal structural part Download PDFInfo
- Publication number
- CN113888446A CN113888446A CN202111279562.0A CN202111279562A CN113888446A CN 113888446 A CN113888446 A CN 113888446A CN 202111279562 A CN202111279562 A CN 202111279562A CN 113888446 A CN113888446 A CN 113888446A
- Authority
- CN
- China
- Prior art keywords
- pixel
- pixels
- line segment
- same
- line
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 58
- 239000002184 metal Substances 0.000 title claims abstract description 54
- 238000005452 bending Methods 0.000 title claims abstract description 33
- 238000000034 method Methods 0.000 claims abstract description 32
- 238000010586 diagram Methods 0.000 claims abstract description 11
- 238000012216 screening Methods 0.000 claims abstract description 9
- 238000003708 edge detection Methods 0.000 claims abstract description 7
- 230000000737 periodic effect Effects 0.000 claims description 25
- 230000001105 regulatory effect Effects 0.000 claims description 2
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000004458 analytical method Methods 0.000 description 3
- 238000005259 measurement Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000010009 beating Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the field of straight line detection, in particular to an intelligent detection method for a sheet metal structural part bending line. The method comprises the steps of obtaining a sheet metal structural part unfolding part, and carrying out edge detection on the sheet metal structural part unfolding part to obtain an edge characteristic diagram; constructing a neighborhood chain code with a direction, and acquiring an extending direction value of a pixel; judging whether the pixels are pixels on the same connected line segment according to the extending direction values of the pixels in the edge feature map, and integrating the pixels on the same connected line segment into a pixel set of the same connected line segment; comparing the number of pixels of each line segment with a set threshold, screening to obtain a suspected dotted line picture in the edge feature graph, and performing linear detection on the suspected dotted line picture to obtain a same linear line segment set; and cutting the straight line segments to obtain independent bent lines, and finishing the detection of the bent lines. According to the invention, the dotted line pixels are segmented by using the dotted line characteristics of the bending lines in the sheet metal unfolded drawing, and only the dotted line pixels are subjected to linear detection, so that the detection efficiency is improved, and meanwhile, the detection result is more accurate.
Description
Technical Field
The invention relates to the field of straight line detection, in particular to an intelligent detection method for a sheet metal structural part bending line.
Background
In the production of sheet metal structure, because some errors may cause the scheduling problem of losing of sheet metal structure drawing, so need recover sheet metal structure drawing according to existing sheet metal structure. The method adopted at present is to measure the sheet metal structural part and then draw the drawing again, so that the method has the disadvantages of long consumed time, overlarge labor cost and possibility of measurement errors in the manual measurement process; or reducing the line segment in the sheet metal structural part by adopting Hough transform, unfolding the sheet metal structural part to obtain an unfolded image of the sheet metal structural part in the process of restoration, and then carrying out Hough transform on the unfolded image of the sheet metal structural part to obtain a restored drawing; however, in the process of unfolding the sheet metal structural part, the bending area is wide, the line segment of the bending part obtained after Hough transform is wide, and the recovered drawing error is large; the size of the bent part cannot be accurately given, and great inconvenience is brought to the post-production of the sheet metal structural part.
Disclosure of Invention
The invention provides an intelligent detection method for a sheet metal structural part bending line, which aims to solve the problem that the detected bending line is wide due to the fact that the bending area of an unfolded sheet metal structural part is wide, and the accurate bending line cannot be obtained, so that inconvenience is caused in the post-production process.
The intelligent detection method for the bending line of the sheet metal structural part adopts the following technical scheme:
the method comprises the following steps:
obtaining a sheet metal structural part unfolding part, and carrying out edge detection on the sheet metal structural part unfolding part to obtain an edge characteristic diagram;
constructing a neighborhood chain code with a direction, and obtaining extension direction values of all pixels in the edge characteristic graph by using the constructed neighborhood chain code;
judging pixels belonging to the same connected line segment in the connected domain pixels by using the extending direction values of the pixels, and integrating the pixels on the same connected line segment into the pixel set of the same connected line segment;
comparing the length of each connected line segment pixel set with a set threshold value, screening out a suspected dotted line area to obtain a suspected dotted line picture in an edge feature graph, carrying out Hough line detection on the suspected dotted line picture, and collecting line segments belonging to the same straight line into the same straight line segment set;
and periodically segmenting the pixel set of the same straight line segment by utilizing the interval of the segment to obtain an independent dotted line segment, and completing the dotted line detection of the edge characteristic graph.
The method for determining the extending direction value of the pixel comprises the following steps: acquiring chain code values of two adjacent pixels of the pixel, wherein the chain code values of the two adjacent pixels of the pixel are extension direction values of the pixel; if a pixel has only a single extension direction, the extension direction value side of the pixel is 0.
When the extension direction values of the connected domain pixels are equal, or the extension direction values of the pixels periodically change at intervals, the connected domain pixels belong to the pixels on the same connected line segment.
When the extending direction values of the pixels are equal, the method for acquiring the pixel sets with the same extending direction values and belonging to the same connected line segment is as follows:
when the extending direction values of the pixels are equal, the pixels adjacent to the pixels in the extending direction extend in the same direction, the relationship of the extending direction values of the adjacent pixels is continuously judged, and if the extending direction values of the adjacent pixels are also equal, the pixels adjacent to the pixels in the extending direction also extend in the same direction;
and repeating the method until non-equidirectionally extending pixels appear or a connected endpoint is reached, and integrating the extending direction values of the obtained equidirectionally extending pixels to obtain a same connected line segment pixel set.
When the extension direction value of the pixel changes periodically, the method for acquiring the pixel set of the same connected line segment with the extension direction value changing periodically comprises the following steps:
extracting single-side extension direction values in pixels of the same connected domain, arranging the single-side extension direction values according to adjacent pixel features to obtain an extension direction sequence, and finding out all periodic sequence values in the sequence; if all the periodic sequence values in the sequence are periodically changed, the pixels are pixels on the same connected line segment, and the extracted pixels of the same connected domain are integrated to obtain a pixel set of the same connected line segment.
When the extending direction value of the pixel is periodically changed at intervals, the method for acquiring the pixel set of the same connected line segment, in which the extending direction value is periodically changed at intervals, comprises the following steps:
finding out continuous periodic sequence pixel sets and non-periodic pixels in the connected domain pixel sets;
judging the position of the non-periodic pixel from the connected domain boundary: the non-periodic pixel and the connected domain boundary are within a certain distance, and cannot form a complete period under the influence of less pixel points, and the pixel and the adjacent pixel are integrated into the same connected line segment pixel set;
calculating the distance between the pixel value and the boundary of the connected domain, and if the obtained distance is less than the number of single period values, the pixel is a boundary pixel, and the pixel and the adjacent pixel belong to the same connected line segment pixel set;
acquiring an aperiodic pixel which does not belong to a connected domain boundary, judging the period sequence values of two adjacent pixels of the aperiodic pixel which does not belong to the connected domain boundary, and judging whether the two period sequence values are the same;
when the two period sequence values are the same, the pixel and the pixels of the left and right line segments belong to the same connected line segment pixel set, and the pixel points of the two line segments are integrated into the same connected line segment pixel set;
when the two periodic sequence values are different, the pixel is the intersection point of the two line segments, and the pixel is discarded;
and integrating the obtained line segment sequences into a connected line segment set, and counting the number of pixels of each connected line segment set.
The process of periodically segmenting the same straight line segment pixel set by using the interval of the line segment is as follows:
calculating the distance between the end points of two adjacent line segments in the same straight line segment set, and screening out the minimum end point distance; obtaining the mode of the minimum endpoint distance; obtaining a broken line segment set by comparing the set distance threshold value with the set distance threshold value; and regulating the line segments with the same interval distance into the same dotted line segment set to obtain a plurality of independent dotted line segment areas, and finishing dotted line detection.
The invention has the beneficial effects that: according to the method, before straight line detection is carried out by Hough transform, the dotted line area in the drawing is divided by using the dotted line characteristic of the bending line, and only the dotted line area is subjected to straight line detection, so that the workload of straight line detection is reduced, the detection efficiency of the bending line is improved, the detected bending line area is thinner, and the drawing of the sheet metal structural part can be restored more accurately.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
FIG. 1 is a schematic flow chart of a method according to an embodiment of the present invention;
FIG. 2 is a sheet metal structural member deployment piece in this embodiment;
FIG. 3 is a schematic table of chain codes constructed in the present embodiment;
FIG. 4 is a diagram illustrating the extending direction values of the whole connected component pixels in this embodiment;
FIG. 5 is a schematic diagram of stepped line segment pixels according to the present embodiment;
fig. 6 is a schematic diagram of the present embodiment showing the same set of straight line segments but with different dashed line segments.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The embodiment of the intelligent detection method for the sheet metal structural part bending line, as shown in fig. 1 to 6, is described below by combining the embodiment with the drawings:
the invention mainly aims to divide dotted line pixels by using the dotted line characteristics of the sheet metal structural member bending line and then perform linear detection on the dotted line pixels when detecting the sheet metal structural member bending line, so that the workload of linear detection is reduced and the detection efficiency is improved.
The method comprises the following steps:
obtaining a sheet metal structural part unfolding part, and carrying out edge detection on the sheet metal structural part unfolding part to obtain an edge characteristic diagram;
constructing a neighborhood chain code with a direction, and obtaining extension direction values of all pixels in the edge characteristic graph by using the constructed neighborhood chain code;
judging pixels belonging to the same connected line segment in the connected domain pixels by using the extending direction values of the pixels, and integrating the pixels on the same connected line segment into the pixel set of the same connected line segment;
comparing the length of each connected line segment pixel set with a set threshold value, screening out a suspected dotted line area to obtain a suspected dotted line picture in an edge feature graph, carrying out Hough line detection on the suspected dotted line picture, and collecting line segments belonging to the same straight line into the same straight line segment set;
and periodically segmenting the pixel set of the same straight line segment by utilizing the interval of the segment to obtain an independent dotted line segment, and completing the dotted line detection of the edge characteristic graph.
1. And obtaining a sheet metal structural part unfolding part, and carrying out edge detection on the sheet metal structural part unfolding part to obtain an edge characteristic diagram.
According to the embodiment, the sheet metal structural part bending line needs to be detected according to the sheet metal structural part unfolding part, so that the sheet metal structural part unfolding part in the JPG format needs to be obtained, and then the sheet metal structural part unfolding part is subjected to edge detection to obtain an edge characteristic diagram.
Obtaining a sheet metal structural part unfolding piece: and (3) unfolding the three-dimensional image of the window into a two-dimensional plane image in a JPG format by utilizing SolidWorks software, and manually uploading the image of each visual angle to a sheet metal structural part processing system, so that the sheet metal structural part unfolded part of each visual angle can be obtained.
Carrying out edge detection on the sheet metal structural member unfolded part to obtain an edge characteristic diagram: and processing the sheet metal structural part unfolded part by using a canny operator to obtain an edge image of the sheet metal structural part unfolded part.
2. Constructing a neighborhood chain code with a direction, analyzing the extending direction of pixels to obtain a plurality of connected line segment pixel sets, comparing the length of each connected pixel line segment with a set threshold value, screening out a suspected dotted line region picture in an edge feature graph, and performing linear detection on the suspected dotted line picture to obtain the same linear line segment set.
In this embodiment, before performing the line detection, a suspected dotted line pixel is obtained by using a dotted line feature, then a dotted line pixel point set is obtained by performing the line detection on the suspected dotted line pixel, and then a bending line is obtained through an interval period of a dotted line.
The dashed line features include the length of the connected line segments, which are relatively small.
At present, the line detection of bending of the sheet metal structural member unfolding piece is generally performed by firstly performing line detection through Hough transform, and then performing broken line judgment on the basis of line detection. This way, a large amount of unnecessary workload is increased, that is, each pixel point in the solid line area is subjected to straight line verification. When the line detection is carried out, the sine curve corresponding to each pixel is obtained according to the line passing through each direction of each pixel, and the line detection is realized according to the number of the intersection points of the sine curve. Therefore, the workload of the linear detection of each pixel point is increased by several times, and the detection efficiency of the dotted line detection is low by the method.
In this embodiment, an extending direction value of a pixel is obtained according to a constructed neighborhood chain code value, a connected line segment is obtained by using the extending direction value of the pixel, and a suspected dotted line is obtained by thresholding the length of a pixel set of the connected line segment, which includes the following specific processes:
2.1 neighborhood analysis obtains the extension direction value of the pixel:
2.1.1 construction of neighborhood chain code:
and constructing neighborhood chain codes with directions, and judging whether the connected domain pixels are continuously connected in a fixed direction or not, namely the neighborhood chain codes of the pixels on the same straight line are the same. The center of the neighborhood chain code is the position of the central pixel, the horizontally adjacent pixels are in the 1 direction, the vertically adjacent pixels are in the 3 direction, the diagonally adjacent pixels are in the 2 direction and the 4 direction respectively, and the specific chain code is shown in fig. 3.
2.1.2 rules of movement of neighborhood chain codes
Judging the pixels by using the neighborhood chain code
After the extension direction of the neighborhood, the chain code needs to be moved to the adjacent pixels at two sides, the extension direction of the adjacent pixel is judged, all connected pixels are traversed through the moving mode, and the extension direction value of the pixel of the whole connected domain is obtained
. As shown in particular in fig. 4.
2.1.3 determine the extending direction of the adjacent pixels of the line by using the neighborhood chain code:
selecting a pixel point on the line edge pixel
Judging the extending direction of adjacent pixels by using the neighborhood chain code value, calculating the extending direction of the pixels in two extending directions, and recording the extending direction value of the pixels in the two extending directions as
. When in use
When the pixel is a pixel, the adjacent pixels on the two sides of the pixel extend in the same direction. When the pixel has only a single extension direction, the extension direction value is
。
2.2 according to the change rule of the extension direction, integrating the line segment set:
and acquiring the same connected line segment pixel set belonging to the same line segment pixel by using the extension direction value.
2.2.1 if the extension direction values are the same (horizontal straight line and vertical straight line), obtaining the pixel set of the same connected line segment by the following method:
judging whether the pixel is extended in the same direction by using the extending direction value when the extending direction value of the pixel is
When the pixel is adjacent to the pixel, the pixel extends in the same direction;
then, the extension direction values of the adjacent pixels are judged
If the extension direction values are equal, the pixels are the pixels extending in the same direction, the judgment is continued according to the rule until the pixels extending in the different directions are found or the connected end point is reached, and the pixel set extending in the same direction can be obtained through the method
. The pixels in the set are necessarily pixels on the same line segment, and the pixels are integrated into the same connected line segment pixel set.
2.2.2 only partial line segment detection can be realized by the above method, however, the following situation exists in real life, as shown in fig. 5;
if the extension direction value is periodically changed or periodically changed at intervals, acquiring the same connected line segment pixel set belonging to the same line segment pixel by using the following method:
the pixels shown in fig. 5 do not extend in the same direction, but they also belong to pixels on the same line segment, so for this case, the variation rule of the pixel extending direction needs to be further analyzed to obtain the same connected line segment pixel set belonging to the same line segment pixel.
As can be seen from fig. 5, although the extending direction values of the pixels belonging to the same connected line segment change, the extending direction values thereof periodically or intermittently periodically change; the change rule of the extending direction of the pixels on different line segments after the intersection point of the line segments is different from that of the pixels on the same line segment.
(1) Assuming the extension of all pixels in the same connected domainValue of extension direction
Extracting one-side extension direction value, arranging the values according to the adjacent features of the pixels to obtain an extension direction sequence
Analyzing the periodicity of the value variation of the sequence to find all the periodic sequence values in the sequence, such as the linear image shown in FIG. 5, in which only one of the pixel extension direction values is [4, 1, 4 ]]。
(2) Analyzing the periodicity of connected pixels in the connected domain pixel set:
finding consecutive sets of periodic sequence pixels in a set of connected component pixels
And non-periodic pixels
；
(a) Determining boundary pixels in the aperiodic pixels:
judging the position of the non-periodic pixel from the connected domain boundary: when the pixel is located at a position closer to the boundary of the connected domain, the extending direction value of the pixel is not periodic due to the influence of the small number of the pixel points, and cannot form a complete period, so that the pixel and the adjacent pixels are integrated into a connected line segment set. The pixel is considered as a boundary pixel.
Calculating the distance of the pixel value from the connected component boundary
When distance is exceeded
When the pixel is a boundary pixel, the pixel and the adjacent pixel belong to the same line segment set. Here, the
Taking the value as the number of pixels in a single period.
(b) Analyzing the relation between non-periodic pixels of non-boundary pixels and the periodic sequence pixel set:
analyzing the periodic consistency rule of adjacent pixels at two sides of the non-periodic pixel: acquiring a period sequence value to which pixels of adjacent line segments on two sides of a non-periodic pixel which does not belong to a boundary pixel belong
Judging whether the two periodic sequence values are the same;
when in use
When the pixel period sequence values of the line segments on the two sides are the same, the pixel and the pixels of the line segments on the left side and the right side belong to the same pixel set, namely the two line segments
Integration into the same connected line segment pixel set
Performing the following steps;
when in use
And if the pixel period sequence values of the line segments on the two sides are different, the pixel is the intersection point of the two line segments, and the pixel is discarded.
By the method, the small line segment sequences are integrated into a plurality of connected line segment pixel sets
And counting the number of pixels of each connected line segment pixel set.
2.3 thresholding the maximum connected length to obtain the suspicious dashed region:
comparing the length of each connected line segment pixel set with a set threshold value to screen out a suspected dotted line area
The number of the corresponding elements in each line segment set is
When it comes to
Each line segment is a suspected dashed line region. Here, the
Is set according to the length of the broken line segment.
By the method, the dotted line pixel area can be segmented, and the suspected dotted line picture can be obtained.
The steps carry out detailed analysis on all edge lines in the sheet metal structural part unfolding piece before the straight line detection is carried out, a suspected dotted line area is determined more accurately, misjudgment of non-bending lines when straight line detection is carried out by directly using Hough change is reduced, and efficiency and accuracy of sheet metal structural part bending line detection are improved.
2.4, carrying out straight line detection by using Hough transform to obtain a same straight line segment set:
carrying out straight line detection on the suspected dotted line picture by using Hough transform, and integrating line segment sets on the same straight line into one set to obtain the same straight line segment set
。
3. And further segmenting the pixel set of the same straight line segment periodically according to the interval of the segment to obtain an independent dotted line segment, and completing the dotted line detection of the edge characteristic graph.
Although the line segment sets on the same straight line can be integrated into the same straight line segment set in the step two during the straight line detection, there may be line segments belonging to the same straight line segment set but belonging to different dotted line segments, as shown in fig. 6. In order to cope with this situation, it is necessary to periodically divide the straight line further according to the interval of the line segment to obtain an independent broken line segment. The specific process is as follows:
3.1 calculating the distance between the end points of two adjacent line segments in the same straight line segment set:
Because there is a space between two connected line segments in the dotted line, the distance between two adjacent end points needs to be calculated; and the length of each connected line segment is smaller, so the distance of the screening end point is the minimum.
3.2 statistical end-point distance mode:
mode of statistical line segment end point distance
。
Among all the obtained end point distances, the distance with the largest occurrence frequency is the distance of each connecting line segment of the dotted line, so the mode of the end point distance of the outgoing line segment needs to be counted.
3.4 comparing the distance threshold to obtain a broken line segment set:
comparing the distance between the end points with a set threshold value when
In time, it is shown that the two line segments do not belong to the same dashed line segment.
And 3.4, by means of the line segment spacing distance, the line segments with the same line segment spacing distance are classified into the same virtual line segment set, and a plurality of independent virtual line segment areas can be obtained in the mode, so that the detection of the virtual lines of the edge characteristic diagram is completed.
After utilizing hough transform to carry out the straight line detection, according to the dotted line characteristic again (because the line of bending is generally the form of dotted line and marks out on the drawing, so the detection of this embodiment all is based on the basis that the dotted line detected, this embodiment the dotted line is the line of bending promptly), carry out further analysis and verification for the line of bending that detects out can not be very wide, thereby make the drawing that restores more accurate, can guarantee that the later stage is more convenient in the production process to the panel beating structure spare.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (7)
1. The intelligent detection method for the bending line of the sheet metal structural part is characterized by comprising the following steps of: the method comprises the following steps:
obtaining a sheet metal structural part unfolding part, and carrying out edge detection on the sheet metal structural part unfolding part to obtain an edge characteristic diagram;
constructing a neighborhood chain code with a direction, and obtaining extension direction values of all pixels in the edge characteristic graph by using the constructed neighborhood chain code;
judging pixels belonging to the same connected line segment in the connected domain pixels by using the extending direction values of the pixels, and integrating the pixels on the same connected line segment into the pixel set of the same connected line segment;
comparing the length of each connected line segment pixel set with a set threshold value, screening out a suspected dotted line area to obtain a suspected dotted line picture in an edge feature graph, carrying out Hough line detection on the suspected dotted line picture, and collecting line segments belonging to the same straight line into the same straight line segment set;
and periodically segmenting the pixel set of the same straight line segment by utilizing the interval of the segment to obtain an independent dotted line segment, and completing the dotted line detection of the edge characteristic graph.
2. The intelligent detection method for the sheet metal structure bending line according to claim 1, characterized in that: the method for determining the extending direction value of the pixel comprises the following steps: acquiring chain code values of two adjacent pixels of the pixel, wherein the chain code values of the two adjacent pixels of the pixel are extension direction values of the pixel; if a pixel has only a single extension direction, the extension direction value side of the pixel is 0.
3. The intelligent detection method for the sheet metal structure bending line according to claim 1, characterized in that: when the extension direction values of the connected domain pixels are equal, or the extension direction values of the pixels periodically change at intervals, the connected domain pixels belong to the pixels on the same connected line segment.
4. The intelligent detection method for the sheet metal structure bending line according to claim 3, characterized in that: when the extending direction values of the pixels are equal, the method for acquiring the same connected line segment pixel set with the same extending direction value is as follows:
when the extending direction values of the pixels are equal, the pixels adjacent to the pixels in the extending direction extend in the same direction, the relationship of the extending direction values of the adjacent pixels is continuously judged, and if the extending direction values of the adjacent pixels are also equal, the pixels adjacent to the pixels in the extending direction also extend in the same direction;
and repeating the method until non-equidirectionally extending pixels appear or a connected endpoint is reached, and integrating the extending direction values of the obtained equidirectionally extending pixels to obtain a same connected line segment pixel set.
5. The intelligent detection method for the sheet metal structure bending line according to claim 3, characterized in that: when the extension direction value of the pixel changes periodically, the method for acquiring the same connected line segment pixel set with the extension direction value changing periodically comprises the following steps:
extracting single-side extension direction values in pixels of the same connected domain, arranging the single-side extension direction values according to adjacent pixel features to obtain an extension direction sequence, and finding out all periodic sequence values in the sequence; if all the periodic sequence values in the sequence are periodically changed, the pixels are pixels on the same connected line segment, and the extracted pixels of the same connected domain are integrated to obtain a pixel set of the same connected line segment.
6. The intelligent detection method for the sheet metal structure bending line according to claim 3, characterized in that: when the extension direction value of the pixel changes periodically at intervals, the method for acquiring the same connected line segment pixel set with the extension direction value changing periodically at intervals comprises the following steps:
finding out continuous periodic sequence pixel sets and non-periodic pixels in the connected domain pixel sets;
judging the position of the non-periodic pixel from the connected domain boundary: the non-periodic pixel and the connected domain boundary are within a certain distance, and cannot form a complete period under the influence of less pixel points, and the pixel and the adjacent pixel are integrated into the same connected line segment pixel set;
calculating the distance between the pixel value and the boundary of the connected domain, and if the obtained distance is less than the number of single period values, the pixel is a boundary pixel, and the pixel and the adjacent pixel belong to the same connected line segment pixel set;
acquiring an aperiodic pixel which does not belong to a connected domain boundary, judging the period sequence values of two adjacent pixels of the aperiodic pixel which does not belong to the connected domain boundary, and judging whether the two period sequence values are the same;
when the two period sequence values are the same, the pixel and the pixels of the left and right line segments belong to the same connected line segment pixel set, and the pixel points of the two line segments are integrated into the same connected line segment pixel set;
when the two periodic sequence values are different, the pixel is the intersection point of the two line segments, and the pixel is discarded;
and integrating the obtained line segment sequences into a connected line segment set, and counting the number of pixels of each connected line segment set.
7. The intelligent detection method for the sheet metal structure bending line according to claim 1, characterized in that: the process of periodically segmenting the same straight line segment pixel set by using the interval of the line segment is as follows:
calculating the distance between the end points of two adjacent line segments in the same straight line segment set, and screening out the minimum end point distance; obtaining the mode of the minimum endpoint distance; obtaining a broken line segment set by comparing the set distance threshold value with the set distance threshold value; and regulating the line segments with the same interval distance into the same dotted line segment set to obtain a plurality of independent dotted line segment areas, and finishing dotted line detection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111279562.0A CN113888446B (en) | 2021-11-01 | 2021-11-01 | Intelligent detection method for bending line of sheet metal structural part |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111279562.0A CN113888446B (en) | 2021-11-01 | 2021-11-01 | Intelligent detection method for bending line of sheet metal structural part |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113888446A true CN113888446A (en) | 2022-01-04 |
| CN113888446B CN113888446B (en) | 2024-10-25 |
Family
ID=79014668
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111279562.0A Active CN113888446B (en) | 2021-11-01 | 2021-11-01 | Intelligent detection method for bending line of sheet metal structural part |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113888446B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113706535A (en) * | 2021-10-28 | 2021-11-26 | 江苏启力锻压机床有限公司 | Sheet metal part bending notch detection method and system based on image processing |
| CN116721101A (en) * | 2023-08-09 | 2023-09-08 | 山东华航高分子材料有限公司 | Defect detection method for surface wave flow marks of plastic part |
| CN117095004A (en) * | 2023-10-20 | 2023-11-21 | 金成技术股份有限公司 | Excavator walking frame main body welding deformation detection method based on computer vision |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109801281A (en) * | 2019-01-23 | 2019-05-24 | 浙江大学 | A kind of texture-free metal parts image regulation straightway edge detection method |
| CN110363200A (en) * | 2019-06-12 | 2019-10-22 | 江苏富山软件科技有限公司 | A kind of line segment detecting method and its system |
| WO2020147398A1 (en) * | 2019-01-17 | 2020-07-23 | 平安科技（深圳）有限公司 | Reproduced image detection method and device, computer device and storage medium |
| CN112862847A (en) * | 2021-03-12 | 2021-05-28 | 辽宁工程技术大学 | Line segment sequence detection method for expressing surface feature edge |
| CN112991283A (en) * | 2021-03-04 | 2021-06-18 | 华南理工大学 | Flexible IC substrate line width detection method based on super-pixels, medium and equipment |
-
2021
- 2021-11-01 CN CN202111279562.0A patent/CN113888446B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020147398A1 (en) * | 2019-01-17 | 2020-07-23 | 平安科技（深圳）有限公司 | Reproduced image detection method and device, computer device and storage medium |
| CN109801281A (en) * | 2019-01-23 | 2019-05-24 | 浙江大学 | A kind of texture-free metal parts image regulation straightway edge detection method |
| CN110363200A (en) * | 2019-06-12 | 2019-10-22 | 江苏富山软件科技有限公司 | A kind of line segment detecting method and its system |
| CN112991283A (en) * | 2021-03-04 | 2021-06-18 | 华南理工大学 | Flexible IC substrate line width detection method based on super-pixels, medium and equipment |
| CN112862847A (en) * | 2021-03-12 | 2021-05-28 | 辽宁工程技术大学 | Line segment sequence detection method for expressing surface feature edge |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113706535A (en) * | 2021-10-28 | 2021-11-26 | 江苏启力锻压机床有限公司 | Sheet metal part bending notch detection method and system based on image processing |
| CN113706535B (en) * | 2021-10-28 | 2022-03-15 | 江苏启力锻压机床有限公司 | Sheet metal part bending notch detection method and system based on image processing |
| CN116721101A (en) * | 2023-08-09 | 2023-09-08 | 山东华航高分子材料有限公司 | Defect detection method for surface wave flow marks of plastic part |
| CN116721101B (en) * | 2023-08-09 | 2023-10-20 | 山东华航高分子材料有限公司 | Defect detection method for surface wave flow marks of plastic part |
| CN117095004A (en) * | 2023-10-20 | 2023-11-21 | 金成技术股份有限公司 | Excavator walking frame main body welding deformation detection method based on computer vision |
| CN117095004B (en) * | 2023-10-20 | 2024-01-12 | 金成技术股份有限公司 | Excavator walking frame main body welding deformation detection method based on computer vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113888446B (en) | 2024-10-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113888446A (en) | Intelligent detection method for bending line of sheet metal structural part | |
| CN106960208B (en) | Method and system for automatically segmenting and identifying instrument liquid crystal number | |
| US8660349B2 (en) | Screen area detection method and screen area detection system | |
| CN114494017B (en) | Method, device, equipment and medium for adjusting DPI (deep packet inspection) image according to scale | |
| CN108921813B (en) | Unmanned aerial vehicle detection bridge structure crack identification method based on machine vision | |
| CN109376740A (en) | A kind of water gauge reading detection method based on video | |
| CN101995412B (en) | Robust glass scratch defect detection method and device thereof | |
| CN115082462A (en) | Method and system for detecting appearance quality of fluid conveying pipe | |
| CN114820625B (en) | Automobile top block defect detection method | |
| CN106778754A (en) | A kind of industrial ammeter digit recognition method of robust | |
| CN103310435A (en) | Method for partitioning number plate characters by combining vertical projection and optimal path | |
| CN106530273B (en) | High-precision FPC (Flexible printed Circuit) linear line detection and defect positioning method | |
| CN114140384A (en) | Transverse vibration image recognition algorithm for hoisting steel wire rope based on contour fitting and centroid tracking | |
| CN114119466A (en) | Method, device and equipment for detecting surface defects of aluminum foil and computer storage medium | |
| CN116993742A (en) | Nickel alloy rolling defect detection method based on machine vision | |
| CN112508913A (en) | Cable section edge detection method based on image detection | |
| CN108764343B (en) | Method for positioning tracking target frame in tracking algorithm | |
| CN111476792B (en) | Extraction method of strip steel image contour | |
| CN114581890B (en) | Method and device for determining lane line, electronic equipment and storage medium | |
| CN111161264A (en) | Method for segmenting TFT circuit image with defects | |
| CN116596921A (en) | Method and system for sorting incinerator slag | |
| CN111860521B (en) | Method for segmenting distorted code-spraying characters layer by layer | |
| CN115170561A (en) | Roundness detection method and system of laser cutting workpiece based on visual processing | |
| CN114842041A (en) | Embedded-based steel rail profile detection light bar extraction method | |
| CN113920121A (en) | Artwork printing streak detection method and system based on image processing |
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 |