CN111091537B - Method for evaluating full rolling process simulation hole pattern fullness - Google Patents
Method for evaluating full rolling process simulation hole pattern fullness Download PDFInfo
- Publication number
- CN111091537B CN111091537B CN201911190496.2A CN201911190496A CN111091537B CN 111091537 B CN111091537 B CN 111091537B CN 201911190496 A CN201911190496 A CN 201911190496A CN 111091537 B CN111091537 B CN 111091537B
- Authority
- CN
- China
- Prior art keywords
- section
- rolled piece
- fullness
- roller
- pixel
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30116—Casting
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the field of steel rolling production, and discloses a full rolling process simulation hole pattern fullness evaluation method which is used for automatically giving quantitative indexes of hole pattern fullness in a full rolling process simulation report so as to reduce the workload. Firstly, extracting a hole type section picture and a rolled piece section picture at a rolling line from a full rolling process simulation picture, then converting the extracted pictures into gray images, and then binarizing the images into black and white images by using a pixel threshold judgment method; then extracting the contour lines of the hole-shaped section and the rolled piece section respectively, and obtaining a rolled piece section diagram based on the contour lines of the rolled piece section; then extracting the section of each roller in the pass outline drawing; fitting a pass closed graph based on the cross section of each roller; then, respectively calculating the area S1 of the hole-type closed diagram and the area S2 of the rolled piece section diagram; and finally, calculating the ratio of the area S2 to the area S1, wherein the ratio is the full rolling process simulation hole pattern fullness. The method is suitable for evaluating the full-rolling-process simulation hole pattern fullness.
Description
Technical Field
The invention relates to the field of steel rolling production, in particular to an evaluation method of full-rolling-process simulation hole pattern fullness.
Background
The pass design is a key link of the whole section rolling, the size of the metal filling degree inside the pass determines whether a key index meeting the specification of a product can be rolled, at present, a simulation technology is generally adopted to test the pass design, whether the pass design is reasonable or not is observed from a simulation result, after the simulation result is obtained, the fullness index is obtained by manually extracting the section of a rolled piece and then comparing the section with the area of a pass filling area, the process is complex, the labor intensity is high, and the integration of the simulation data of the whole rolling process is not facilitated.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the evaluation method of the full rolling process simulation hole pattern fullness is provided and used for automatically giving quantitative indexes of the hole pattern fullness in a full rolling process simulation report so as to reduce the workload.
In order to solve the problems, the invention adopts the technical scheme that: a full rolling process simulation hole pattern fullness evaluation method comprises the following steps:
step 1: extracting a hole pattern section picture and a rolled piece section picture at a rolling line from the full rolling process simulation picture, converting the extracted pictures into gray images, and then binarizing the images into black and white images by using a pixel threshold judgment method;
And 2, step: respectively extracting contour lines of the hole type section and the rolled piece section from the hole type section picture and the rolled piece section picture after binarization, and obtaining a rolled piece section diagram based on the contour lines of the rolled piece section;
and 3, step 3: extracting the section of each roller in the hole pattern profile diagram;
and 4, step 4: fitting a pass closed graph based on the cross section of each roller;
and 5: respectively calculating the area S1 of the hole type closed diagram and the area S2 of the rolled piece section diagram;
step 6: and calculating the ratio of the area S2 to the area S1, wherein the ratio is the full rolling process simulation hole pattern fullness.
Further, step 2 may use a sobel algorithm to extract the contour lines. The Sobel operator has the advantages of simple method, high processing speed and smooth and continuous obtained edges.
Further, the method for extracting each roll section in step 3 can be as follows:
starting from the corner (such as the upper left corner) of the image, an edge contour is searched by adopting an 8-neighborhood mode, namely, a 9-by-9 square matrix is formed by the current pixel point and 8 surrounding adjacent pixel points, the current pixel point is the center of the matrix, from top to bottom, from left to right, whether the pixel of the adjacent point is 1 is traversed, if the pixel is 1, the matrix is changed into a new matrix taking the point as the center, and the steps are repeated in such a way, when the pixels around the pixel at the center of the matrix are all 0, a closed body entity is found, and the closed body entity is a roller section.
Further, step 4 is to find out a pair of closed mark points P for the two adjacent roll sections respectivelyijAnd PjiIn which P isijRepresenting the nearest pixel point on the ith roller section from the jth roller section, PjiRepresenting a pixel point which is closest to the ith roller section on the jth roller section; then changing the pixel value between each pair of closed mark points to 1; and finally, only keeping pixel point values among the mark points on each roll section, and changing the rest into 0 values so as to obtain a pass closed graph.
The invention has the beneficial effects that: the invention can automatically extract the rolled piece section diagram and the pass closing diagram from the full rolling process simulation diagram, and compares the rolled piece section diagram with the pass filling area to obtain the pass fullness, and the whole process is automatically completed through a program, thereby reducing the workload of workers and improving the working efficiency of the system.
Drawings
Fig. 1 is a schematic diagram of 8-neighborhood discrimination.
FIG. 2 is a schematic cross-sectional view of a roll in the example.
FIG. 3 is a schematic diagram of a mark point on the cross section of the roller in the embodiment
FIG. 4 is a schematic flow chart of a fitting pass closure chart according to an embodiment.
Fig. 5 is a schematic flow chart of an embodiment.
Numbering in the figures: the number of the left roller is 1, the number of the upper roller is 2, and the number of the lower roller is 3.
Detailed Description
In order to enable a system to automatically give quantitative indexes of the hole pattern fullness in a full rolling process simulation report so as to reduce the workload, the invention provides an evaluation method of the full rolling process simulation hole pattern fullness, which comprises the following steps:
step 1: and extracting a hole pattern section picture and a rolled piece section picture at a rolling line from the full rolling process simulation picture, converting the extracted pictures into gray images, and binarizing the images into black and white images by using a pixel threshold judgment method.
And 2, step: respectively extracting contour lines of the hole type section and the rolled piece section from the binarized hole type section picture and rolled piece section picture, and obtaining a rolled piece section picture based on the contour lines of the rolled piece section, wherein the contour lines can be extracted by using a sobel algorithm. The Sobel operator has the advantages of simple method, high processing speed and smooth and continuous obtained edges.
And step 3: and extracting the section of each roller in the pass outline drawing. The method for extracting the cross section of each roller can be as follows:
starting from the corner (such as the upper left corner) of the image, an edge contour is searched by adopting an 8-neighborhood mode, namely, a 9-by-9 square matrix is formed by the current pixel point and 8 surrounding adjacent pixel points, the current pixel point is the center of the matrix, from top to bottom, from left to right, whether the pixel of the adjacent point is 1 is traversed, if the pixel of the adjacent point is 1, the matrix is changed into a new matrix taking the point as the center, and the steps are repeated, so that when the pixels around the pixel at the center of the matrix are all 0, a closed body entity is found, and the closed body entity is a roller section.
And 4, step 4: and fitting a pass closed graph based on the section of each roller. When fitting, a pair of closed mark points P can be found out respectively for the sections of two adjacent rollersijAnd PjiIn which P isijRepresenting the nearest pixel point on the ith roller section from the jth roller section, PjiRepresenting a pixel point which is closest to the ith roller section on the jth roller section; then changing the pixel value between each pair of closed mark points to 1; and finally, only keeping pixel point values among the mark points on each roll section, and changing the rest into 0 values so as to obtain a pass closed graph.
In a specific embodiment, if there are four roll solid sections, the pixel minimum distance between the left roll and the upper and lower rolls can be found out from left to right, if found, the next line is continuously found out, if the line is found out to be smaller than or equal to the previous line, the next line is continuously traversed until the pixel minimum distance of the next line of the line is increased, the line is regarded as an effective contact pass area between the left roll, the upper roll and the rolled piece, two pixel position points corresponding to the distance are recorded, then line changing is continuously carried out to find the pixel minimum distance between the left roll and the lower roll, if the pixel minimum distance of the next line of the line is reduced or not changed, the position is regarded as the effective contact pass area between the left roll, the lower roll and the rolled piece, the two pixel position points corresponding to the distance are recorded, and similarly, the effective contact pass areas between the right roll, the upper roll and the rolled piece are found from right to left; if the solid sections of the three rollers are provided, only the left roller or the right roller is searched in the same process; if only the sections of the upper roller and the lower roller exist, searching effective contact areas of the upper roller and the lower roller and the rolled piece from top to bottom, firstly, searching the pixel minimum distance of the horizontal upper roller and the horizontal lower roller from top to bottom according to rows, if the pixel minimum distance is found, continuously searching the next row, if the pixel minimum distance is found to be smaller than or equal to that of the previous row, continuously traversing the next row until the pixel minimum distance of the next row of the row is increased, considering the effective contact hole type areas of the upper roller and the lower roller and the rolled piece, and recording two pixel position points corresponding to the distance; and finally, changing the pixel values among all the recorded minimum distance positions to be 1, only keeping the pixel point values of the region class determined by the minimum distance positions on the solid section of the roller, and changing the rest values to be 0 values, thereby completing the pass closed graph of the effective contact region.
And 5: respectively calculating the area S1 of the hole type closed diagram and the area S2 of the rolled piece section diagram;
step 6: and calculating the ratio of the area S2 to the area S1, wherein the ratio is the full rolling process simulation hole pattern fullness.
Examples
The invention is further illustrated by the following examples and figures.
The embodiment provides a method for evaluating the full rolling process simulation hole pattern fullness, which is shown in a combined figure 5 and comprises the following steps:
1. and (3) binaryzation of hole pattern and rolled piece section pictures:
and extracting a hole pattern section picture and a rolled piece section picture at a rolling line from the full rolling process simulation graph, converting the extracted pictures into gray images, binarizing the images into black and white images by using a pixel threshold judgment method, and archiving.
2. Contour line extraction:
and respectively extracting the contour lines of the hole type sections and the contour lines of the rolled piece sections from the hole type section pictures and the rolled piece section pictures after binaryzation by adopting a sobel algorithm, obtaining a rolled piece section diagram based on the contour lines of the rolled piece sections, and archiving the rolled piece section diagram.
3. Extracting the section of each roller in the pass profile diagram:
1) all closed faces are searched and defined as entities. Starting from the upper left corner of the image, an 8-neighborhood mode is adopted to search an edge contour, namely, a current pixel point and 8 surrounding adjacent pixel points form a 9-by-9 square matrix, as shown in fig. 1, the current pixel point is the center of the matrix, whether the pixel of the traversing adjacent point is 1 or not is traversed from top to bottom, and from left to right, if the pixel of the traversing adjacent point is 1, the matrix is changed into a new matrix taking the point as the center, the process is repeated in such a way, when all the pixels around the pixel at the center of the matrix are 0, a closed body entity is found, namely a roller section, the pixel information of the closed body is stored, and the next body is continuously searched from the current position until the whole image is traversed.
2) And dividing the entity according to the arrangement direction of the rollers. Extracting horizontal and vertical central lines of an image, using the horizontal and vertical central lines for entity orientation reference, defining a floating depth (generally 10 pixels), calculating the floating times (the number of central line pixels is divided by the floating depth) of two horizontal and longitudinal central lines, floating the horizontal central line from top to bottom, judging whether an entity exists, if only one entity exists, defining the horizontal central line as a lower roll, if two entities exist, defining the horizontal central line as an upper roll, then defining the horizontal central line as a lower roll, if not, traversing the horizontal central line from bottom to top, defining the horizontal central line as a lower roll, then traversing the horizontal central line as an upper roll, if not, defining the upper roll and the lower roll, after traversing from top to bottom, starting floating the vertical central line, and if only one entity exists, defining the horizontal central line as a right roll, and if two entities exist, defining the horizontal central line as a left roll, and defining the subsequent traversal as a right roller, if not, returning the vertical center line to the initial position, traversing from right to left, firstly traversing to the right roller, then traversing to the left roller, and if not, not having the left roller and the right roller, and ending the traversal.
4. Fitting a hole type closed graph:
the hole pattern diagram in fig. 2 is composed of different rolls, and is not a closed diagram, so that the contact surfaces of the divided roll entities in the diagram and the rolled piece which are possibly contacted need to be subjected to closing operation so as to calculate the area of the closed region. In the fitting process, as shown in fig. 3, a pair of closed mark points P on the left roll 1 and the upper roll 2 are respectively found for the two adjacent roll sections12And P21A pair of closed mark points P on the left roller 1 and the lower roller 313And P31A pair of closed mark points P on the upper roller 2 and the lower roller 323And P32Wherein P is12Indicating the nearest pixel point on the section of the left roller 1 to the section of the upper roller 2, P21Represents the nearest pixel point on the section of the upper roller 2 to the section of the left roller 1, P13、P31、P23And P32And so on; then P is added12And P21Between, P12And P21And P23And P32The pixel value in between is changed to 1; only P is remained on the left roller 112And P13Pixel point values in between, leaving only P on the upper roll 221And P23Pixel point values in between, leaving only P on the lower roll 331And P32The pixel point values in between, the rest all become 0 values, so far, a pass closed graph is obtained, and the fitting result is shown in fig. 4.
5. And respectively calculating the area S1 of the hole type closed diagram and the area S2 of the rolled piece section diagram.
6. And calculating the ratio of the area S2 to the area S1, wherein the ratio is the full rolling process simulation hole pattern fullness.
Claims (4)
1. A method for evaluating full rolling process simulation hole pattern fullness is characterized by comprising the following steps:
step 1: extracting a hole pattern section picture and a rolled piece section picture at a rolling line from the full rolling process simulation picture, converting the extracted pictures into gray images, and then binarizing the images into black and white images by using a pixel threshold judgment method;
step 2: respectively extracting contour lines of the hole type section and the rolled piece section from the hole type section picture and the rolled piece section picture after binarization, and obtaining a rolled piece section diagram based on the contour lines of the rolled piece section;
and step 3: extracting the section of each roller in the hole pattern profile diagram;
and 4, step 4: fitting a pass closed graph based on the cross section of each roller;
and 5: respectively calculating the area S1 of the hole type closed diagram and the area S2 of the rolled piece section diagram;
step 6: and calculating the ratio of the area S2 to the area S1, wherein the ratio is the full rolling process simulation hole pattern fullness.
2. The method for evaluating the full rolling process simulation pass fullness as claimed in claim 1, wherein step 2 adopts a sobel algorithm to extract the contour line.
3. The method for evaluating the full pass simulation fullness of a full rolling process as claimed in claim 1, wherein the method for extracting each roll section in step 3 is as follows:
Starting from the corner of the image, an 8-neighborhood mode is adopted to search an edge contour, namely, a 9-by-9 square matrix is formed by the current pixel point and 8 surrounding adjacent pixel points, the current pixel point is the center of the matrix, whether the pixel of the traversing adjacent point is 1 or not is traversed, if the pixel is 1, the matrix is changed into a new matrix taking the point as the center, and the steps are repeated in such a way, when the pixels around the pixel at the center of the matrix are all 0, a closed body entity is found, and the closed body entity is a roller section.
4. The method for evaluating the full-rolling-process simulated pass fullness as claimed in claim 1, wherein step 4 is to find a pair of closed mark points P for two adjacent roll sections respectivelyijAnd PjiIn which P isijRepresenting the nearest pixel point on the ith roller section from the jth roller section, PjiRepresenting a pixel point which is closest to the ith roller section on the jth roller section; then changing the pixel value between each pair of closed mark points to 1; and finally, only keeping pixel point values among the mark points on each roll section, and changing the rest into 0 values so as to obtain a pass closed graph.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911190496.2A CN111091537B (en) | 2019-11-28 | 2019-11-28 | Method for evaluating full rolling process simulation hole pattern fullness |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911190496.2A CN111091537B (en) | 2019-11-28 | 2019-11-28 | Method for evaluating full rolling process simulation hole pattern fullness |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111091537A CN111091537A (en) | 2020-05-01 |
| CN111091537B true CN111091537B (en) | 2022-06-10 |
Family
ID=70393857
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911190496.2A Active CN111091537B (en) | 2019-11-28 | 2019-11-28 | Method for evaluating full rolling process simulation hole pattern fullness |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111091537B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103230940A (en) * | 2013-04-11 | 2013-08-07 | 首钢总公司 | Controlling method for reduction of four-strand slitting line differences and structure property differences |
| CN104077461A (en) * | 2014-07-22 | 2014-10-01 | 攀钢集团攀枝花钢钒有限公司 | Simulation analysis method for full rolling process of steel rail |
| CN104117536A (en) * | 2014-06-24 | 2014-10-29 | 武钢集团昆明钢铁股份有限公司 | Finish-pass-sharing production method of plain round hot-rolled steel bars with diameter being 6.5 mm and disc screw hot-rolled steel bars with diameter being 6 mm |
| CN104820729A (en) * | 2015-03-13 | 2015-08-05 | 攀钢集团攀枝花钢钒有限公司 | Temperature succession method of whole rolling process simulation analysis of steel rail |
| WO2017217246A1 (en) * | 2016-06-17 | 2017-12-21 | 株式会社サタケ | Polishing-type horizontal rice milling machine |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060036419A1 (en) * | 2004-07-29 | 2006-02-16 | Can Technologies, Inc. | System and method for animal production optimization |
-
2019
- 2019-11-28 CN CN201911190496.2A patent/CN111091537B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103230940A (en) * | 2013-04-11 | 2013-08-07 | 首钢总公司 | Controlling method for reduction of four-strand slitting line differences and structure property differences |
| CN104117536A (en) * | 2014-06-24 | 2014-10-29 | 武钢集团昆明钢铁股份有限公司 | Finish-pass-sharing production method of plain round hot-rolled steel bars with diameter being 6.5 mm and disc screw hot-rolled steel bars with diameter being 6 mm |
| CN104077461A (en) * | 2014-07-22 | 2014-10-01 | 攀钢集团攀枝花钢钒有限公司 | Simulation analysis method for full rolling process of steel rail |
| CN104820729A (en) * | 2015-03-13 | 2015-08-05 | 攀钢集团攀枝花钢钒有限公司 | Temperature succession method of whole rolling process simulation analysis of steel rail |
| WO2017217246A1 (en) * | 2016-06-17 | 2017-12-21 | 株式会社サタケ | Polishing-type horizontal rice milling machine |
Non-Patent Citations (2)
| Title |
|---|
| 钢坯缺陷在钢轨轧制过程中的演变规律分析;陶功明等;《轧钢》;20180815;第35卷(第04期);第50-53页 * |
| 钢轨轧制过程典型导卫系统的有限元模拟;原思宇;《塑性工程学报》;20110428;第18卷(第02期);第52-56页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111091537A (en) | 2020-05-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111080620B (en) | Road disease detection method based on deep learning | |
| CN115082467B (en) | Building material welding surface defect detection method based on computer vision | |
| CN113658132B (en) | Computer vision-based structural part weld joint detection method | |
| CN105205488B (en) | Word area detection method based on Harris angle points and stroke width | |
| CN103048329B (en) | A kind of road surface crack detection method based on active contour model | |
| CN107066916B (en) | Scene semantic segmentation method based on deconvolution neural network | |
| CN109215009B (en) | Continuous casting billet surface image defect detection method based on deep convolution neural network | |
| CN102915544A (en) | Video image motion target extracting method based on pattern detection and color segmentation | |
| CN104978567A (en) | Vehicle detection method based on scenario classification | |
| CN111985552B (en) | Method for detecting diseases of thin strip-shaped structure of airport pavement under complex background | |
| CN110210451A (en) | A kind of zebra line detecting method | |
| CN104778458B (en) | A kind of textile pattern search method based on textural characteristics | |
| CN102509109B (en) | Method for distinguishing Thangka image from non-Thangka image | |
| CN113628189A (en) | Rapid strip steel scratch defect detection method based on image recognition | |
| CN109870458B (en) | Pavement crack detection and classification method based on three-dimensional laser sensor and bounding box | |
| CN104637066B (en) | The quick framework extraction method of bianry image based on sequential refinement | |
| CN109685788A (en) | A kind of flooring defect image automatic testing method based on morphological feature | |
| CN109815961B (en) | Pavement repairing type disease detection method based on local texture binary pattern | |
| Liang et al. | An extraction and classification algorithm for concrete cracks based on machine vision | |
| CN113240623A (en) | Pavement disease detection method and device | |
| Chen et al. | The machined surface defect detection of improved superpixel segmentation and two-level region aggregation based on machine vision | |
| CN111091537B (en) | Method for evaluating full rolling process simulation hole pattern fullness | |
| CN107610094B (en) | Three-dimensional crack detection method based on ellipsoid three-dimensional representation | |
| CN105844641A (en) | Adaptive threshold segmentation method in dynamic environment | |
| CN115063817A (en) | Form identification method and system based on morphological detection and storage medium |
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 |