CN102494976B - Method for automatic measurement and morphological classification statistic of ultra-fine grain steel grains - Google Patents
Method for automatic measurement and morphological classification statistic of ultra-fine grain steel grains Download PDFInfo
- Publication number
- CN102494976B CN102494976B CN201110368047.XA CN201110368047A CN102494976B CN 102494976 B CN102494976 B CN 102494976B CN 201110368047 A CN201110368047 A CN 201110368047A CN 102494976 B CN102494976 B CN 102494976B
- Authority
- CN
- China
- Prior art keywords
- image
- grain
- algorithm
- ultra
- steel
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Analysing Materials By The Use Of Radiation (AREA)
Abstract
The invention discloses a method for automatic measurement and morphological classification statistic of ultra-fine grain steel grains. The method provided by the invention comprises the following steps of 1, acquiring an image of an ultra-fine grain steel grain and carrying out pretreatment, 2, carrying out binary segmentation of the pre-treated image by a region division-based self-adaptive threshold segmentation method to obtain a binary image, 3, carrying out grain boundary repair of the binary image by a distance transformation-based modified watershed algorithm, and carrying out grain aperture filling by a modified seed filling algorithm to obtain a repaired image, 4, extracting grain morphology characteristic parameters, and 5, carrying out grading statistic of grain sizes according to diameters, and carrying out grain morphology classification according to roundness, form factors and length-width ratios. Through the method provided by the invention, image repair and accurate and efficient measurement, classification and statistic of an ultra-fine grain steel microstructure (grain) can be realized automatically.
Description
Technical field
The present invention relates to the quantitative metallographic analysis field of ferrous materials microstructure, be specifically related to a kind of automatic measurement and typoiogical classification statistical method thereof of ultra-fine grain crystalline grain of steel.
Background technology
Develop rapidly along with ferrous materials science and technology, the research and development of all kinds of steel have been based upon on the basis of composition, structure, tissue and performance quantitative relationship gradually, thereby meaning is for steel can by preparation and various subsequent technique be controlled its phase structure and microstructure obtains required performance.Quantitative metallographic analysis is studied the important method of relation between metal material composition, tissue, technique and performance just, by the quantitative test to various material metallographic structures, between the microstructure of material and macro property, builds quantitative relationship.
Ultra-fine grain steel is fast-developing in recent years a kind of new type steel, and its principal feature is that its metallographic structure is mostly the extremely crystal grain of refinement, and its crystallite dimension is less than 4 microns conventionally, so show very high intensity, hardness, plasticity and toughness.For ultra-fine grain steel, the particle diameter of its crystal grain, form and distribution play conclusive impact to the performance of steel.When carrying out quantitative metallographic analysis, for a large amount of image deflects such as crystal boundary disappearance, intracrystalline hole that occur in ultra-fine grain steel metallographic structure, must carry out image repair, make original grain boundary true reappearance; Otherwise, by the quantitative metallographic analysis effect having a strong impact on thereafter, cause the quantitative relationship of composition, structure, tissue and the performance of material to be difficult to accurate foundation.In order to improve the performance of ultra-fine grain steel, need to carry out Measurement accuracy, classification and statistics to the particle diameter of ultra-fine grain crystalline grain of steel, form etc.Therefore,, how accurately particle diameter, the form of measurement and statistics crystal grain distribute efficiently, become major issue in the urgent need to address in ultra-fine grain steel Analysis on Microstructure field.
In engineering practice, the main dependence of this work has deep metal material knowledge and the engineering technical personnel that enrich quantitative metallographic analysis experience, and the mode of operation that adopts traditional artificial restorative procedure to carry out metallic phase image repair and conventional mesh method manual measurement, calculating and statistics is carried out measurement, the classification of crystal grain.Therefore the subjective factor that depends primarily on people due to this analytical effect certainly lead to various subjective errors, efficiency low, measure the problem that statistic of classification result precision is low and take in a large number human cost, thereby the quantitative relationship that causes steel product ingredient, structure, tissue and performance is difficult to the consequence of accurately setting up, and this has become " bottleneck " problem that has a strong impact on new material R&D work process.
Summary of the invention
The automatic measurement and the typoiogical classification statistical method thereof that the object of this invention is to provide a kind of ultra-fine grain crystalline grain of steel, metallographic structure (crystal grain) image that the method can realize automatically to ultra-fine grain steel is repaired and its morphological feature is carried out accurately, measures efficiently, classified and statistics.
Technical scheme of the present invention is: a kind of automatic measurement of ultra-fine grain crystalline grain of steel and typoiogical classification statistical method thereof, and its concrete steps are:
(1) gather ultra-fine grain crystalline grain of steel image, and carry out pre-service;
(2) adopt the auto-thresholding algorithm of dividing based on region to carry out binary segmentation to pretreated image, obtain bianry image;
(3) described bianry image is repaired to crystal boundary by the correction watershed algorithm based on range conversion, and fill intracrystalline hole with improving seed fill algorithm, obtain repairing image;
(4) extract grain form characteristic parameter, its concrete steps are:
(4-1) described reparation image is carried out to scale setting and region labeling;
(4-2) extract grain form characteristic parameter: area, girth, length breadth ratio, diameter, circularity and shape coefficient;
(5) take described diameter carries out hierarchical statistics as criterion to crystallite dimension, take described circularity, shape coefficient, length breadth ratio grain form to be classified as criterion;
Further, the pretreated concrete steps of described step (1) are:
(1-1) utilize the histogram equalization algorithm that can retain image detail to strengthen entire image;
(1-2) utilize rim detection method of differential operator to extract edge, the some place of gray scale sudden change is considered as to corresponding frontier point, and then the point set on definite border;
(1-3) utilize stretching algorithm to strengthen the contrast of image simultaneously.
Further, the auto-thresholding algorithm that described step (2) is divided based on region, 2500 of the subregion numerical digits that its region is divided, adopt large Tianjin method algorithm.
Further, the correction watershed algorithm of described step (3) based on range conversion, the steps include:
(3-1) carry out Euclidean Distance Transform, obtain each independent nucleus;
(3-2) according to correction factor, successively expand each independent nucleus, two independent nucleus adhesions after revising, are regarded as an independent nucleus, Unified number;
(3-3) nucleus after described numbering is carried out to expansion process, in expansion process, nucleus keeps increasing with layer position, when two nucleus meet, is watershed divide, now forms the separatrix of crystal grain.
Further, described correction factor is 2.
Advantage of the present invention is:
1, adopt correction watershed segmentation algorithm and improvement seed fill algorithm based on range conversion can solve well respectively the image deflects such as crystal boundary disappearance and intracrystalline hole, can obtain desirable image repair effect.
2, the measuring accuracy of crystal grain image can reach
0.01 μ m, and accomplish without undetected, heavily inspection of nothing.
3, adopt measuring method based on pixel can be accurately, efficiently, carry out easily the measurement classification of grain properties parameter, whole measurement assorting process is moved on the computing machine of standard configuration, the crystal grain that completes a visual field is measured classification only needs a few minutes.
4, the quantitative micro-analysis that the present invention is crystal grain in ultra-fine grain steel provides reliable basis.
5, the present invention has excellent universality, and the shot-like particle that can be applied to all background complexity and complex shape in Material Field, biological field is measured classification work.
Accompanying drawing explanation
Below in conjunction with the drawings and specific embodiments, the present invention is described in further detail:
Fig. 1 FB(flow block) of the present invention;
Fig. 2 is the hardware schematic diagram of image capturing system;
Fig. 3 is the original image of embodiment 1;
Fig. 4 is embodiment 1 image after pretreatment;
Fig. 5 is the image after embodiment 1 binary segmentation;
Fig. 6 (a) is forward direction template; (b) be backward template;
Fig. 7 (a) is the defect crystal grain of crystal boundary to be repaired; (b) be the crystal grain after crystal boundary is repaired;
Fig. 8 (a) is intracrystalline hole image to be filled; (b) be the image of filling after intracrystalline hole;
Fig. 9 is the reparation image of embodiment 1;
Figure 10 (a) is the particle diameter distribution plan of embodiment 1 crystal grain; (b) be the form distribution plan of embodiment 1 crystal grain;
Figure 11 is the original image of embodiment 2;
Figure 12 is the image of embodiment 2 after pre-service and binary segmentation;
Figure 13 is the reparation image of embodiment 2;
Figure 14 (a) is the particle diameter distribution plan of embodiment 2 crystal grain; (b) be the form distribution plan of embodiment 2 crystal grain;
Figure 15 is the original image of embodiment 3;
Figure 16 is the image of embodiment 3 after pre-service and binary segmentation;
Figure 17 is the reparation image of embodiment 3;
Figure 18 (a) is the particle diameter distribution plan of embodiment 3 crystal grain; (b) be the form distribution plan of embodiment 3 crystal grain.
Embodiment
The medium filtering the present invention relates to, contrast stretching algorithm, its particular content is referring to Yang Shuying.VC++ image processing program design (in January, 2005 second edition). publishing house of Tsing-Hua University, the .ISBNB7-81082-450-3/TP.162.PP98-105 of publishing house of Beijing Jiaotong University, 76-80; The histogram equalization algorithm that can retain image detail is that the present invention is at cold fine jade, dawn, Zhang Jiashu. the histogram equalization [J] that jointing edge details local auto-adaptive strengthens. the innovation work on microelectronics and computing machine .2010.1.vol27. No1.PP:38-41. mono-civilian basis; The auto-thresholding algorithm of dividing based on region is that the present invention is auspicious grandson. graphical analysis (in July, 2005 first published). and the innovation work on the .ISBN7-03-013850-3/TP.391.41..PP9-10. of Science Press basis; Correction watershed segmentation algorithm based on range conversion is that the present invention is at J.J. Charles, L.I. Kunchevaa, B. Wells, I.S. Lima.Object segmentation within microscope images of palynofacies[J]. the innovation work on Computers & Geosciences .34 (2008) .PP:688 – 698. one civilian bases.
As shown in Figure 1, first the present invention utilizes image capturing system obtain the original image (target crystal grain image) of crystal grain and deposited in subsidiary image pick-up card.Target crystal grain image is carried out to pre-service, be beneficial to the carrying out of subsequent operation.Only relate to for purposes of the present invention the morphological feature of crystal grain, and irrelevant with colouring information, therefore only need use the adaptive thresholding algorithm of dividing based on region to carry out binary segmentation to it, obtain the black and white template of target image, i.e. the bianry image of crystal grain.Due to crystal grain bianry image " succession " problems such as the peculiar crystal boundary disappearance of original image, intracrystalline hole, also must repair grain boundary by the correction watershed segmentation algorithm by based on distance function, improve seed fill algorithm and fill hole.When completing above-mentioned image repair step and setting after scale, just can carry out region labeling to each crystal grain.Adopt retroactive method and take pixel as measuring unit, target crystal grain being extracted respectively to three initial configuration characteristic parameters: chip area, girth and length breadth ratio; Recycling area, girth, can calculate respectively size of microcrystal, circularity and three characteristic parameters of shape coefficient.
Thus, can according to size of microcrystal, to target image, carry out the hierarchical statistics analysis of size of microcrystal, obtain corresponding analysis diagram;
Then, according to circularity, shape coefficient and length breadth ratio, target image is carried out to typoiogical classification statistics; Finally the automatic classification of above size of microcrystal and form, statistic of classification result are filed and shown output with diagram file form.
Below by 3 embodiment, the present invention is described in detail again:
embodiment 1
Utilize image capturing system to obtain the original grain image of steel, the hardware of image capturing system as shown in Figure 2: steel sample 1, professional microscope 2, camera (CCD) 3, computing machine (interpolated image capture card) 4, printer 5.The concrete steps of image acquisition are to utilize microscope that image is adjusted to proper focal length, make a video recording and store (original image) in image pick-up card into when image is the most clear, can carry out image pre-service.
The original image of embodiment 1 as shown in Figure 3.First the original image of Fig. 3 is carried out to pre-service.First, utilize medium filtering to carry out denoising to image.Then, the histogram equalization algorithm that utilization can retain image detail strengthens processing to entire image, with the detailed information of rich image, thus the display effect of strengthening image.In order further to extract edge, the present invention utilizes rim detection method of differential operator to carry out, and its principle is mainly to utilize the effect of grey scale change.Because its Grad of some place of gray scale sudden change is very high, can be considered corresponding frontier point, thereby determine the point set on border.Utilize stretching algorithm to strengthen the contrast of image, effect after pretreatment as shown in Figure 4 simultaneously.
Fig. 4 is also needed to carry out binary segmentation to obtain the bianry image of crystal grain.Due to diversity, the complicacy of ultra-fine grain crystalline grain of steel image, therefore the present invention adopts the auto-thresholding algorithm of dividing based on region to carry out binary segmentation to image.The auto-thresholding algorithm of dividing based on region presses coordinate piecemeal to image, and each sub-block is obtained respectively to optimal threshold Ti automatically.The present invention finds through lot of experiments, while dividing 2500 sub regions, adopts the large Tianjin of OTSU(method) algorithm, its segmentation effect is best, as shown in Figure 5.
Target image, after above-mentioned pre-service and binary segmentation, improves although its picture quality obtains obviously, but still exists the peculiar defect of ultra-fine grain steel (crystal boundary disappearance, intracrystalline hole etc.), affects the degree of accuracy that target crystal grain is measured classification.For this reason, the present invention improves traditional watershed segmentation algorithm, has formed the new correction watershed segmentation algorithm based on range conversion.Mainly by target image being carried out to range conversion, to obtain the geometric center of each crystal grain be crystal grain core to this algorithm, and each crystal grain core is revised, and to avoid over-segmentation, revised crystal grain image applied to watershed segmentation algorithm again and repair crystal boundary.
The detailed process of the above-mentioned correction watershed segmentation algorithm based on range conversion is: 1. carry out Euclidean Distance Transform, obtain each independent nucleus, its process is: first in bianry image, background gray scale is set to 0, target (crystal grain) gray scale is set to 255, then use forward direction template (as shown in Figure 6 a), backward template (as shown in Figure 6 b) to its from left to right-from top to bottom with from right-to-left-take turns doing from bottom to top twice sweep, when template center arrives a new target location, just the pixel value of each element in template and its correspondence position is added, minimum and value are as the pixel value of current goal.2. according to the size of correction factor, successively expand each independent nucleus, if two independent nucleus adhesions after revising are regarded as an independent nucleus, Unified number.The present invention, according to the characteristics of image of Ultra-fine Grained, shows through repetition test, and adopting correction factor is 2 o'clock, best results.3. the nucleus after above-mentioned numbering is carried out to expansion process, according to the synchronous principle rising of water level, in expansion process, nucleus keeps increasing with layer position, once two nucleus meet, is watershed divide, now forms the separatrix of crystal grain.Fig. 7 a, Fig. 7 b are respectively the grain form after defect grain and grain boundary is repaired.
For intracrystalline hole defect as shown in Figure 8 a, the present invention adopts improved seed fill algorithm to fill processing, and the image after filling as shown in Figure 8 b.This filling algorithm refers to another patent of invention of the inventor: (" the automatic measurement of precipitation particles and typoiogical classification method thereof in a kind of steel ", application number: 200910030216.1).The basic procedure of this seed filling improvement algorithm is as follows:
(1) sub pixel is pressed into storehouse.
(2) when storehouse non-NULL, from storehouse, release a pixel, and this pixel is arranged to desired value.
(3) for each, be communicated with or eight connected pixels with four of current pixel adjacency, test, to determine whether the pixel of test point is in region and not accessed mistake.
(4), if the pixel of testing was not filled in region, this pixel is pressed into storehouse.
In sum, target image has been carried out respectively to pre-service, binary segmentation, crystal boundary reparation and hole and filled after each step process, can obtain the automatic reparation image of view picture target image, as shown in Figure 9.
So far, can carry out measurement, the classification work of crystal grain.First, extract required grain form characteristic parameter, leaching process is:
(1) set image rulers, i.e. the physical size of each pixel in uncalibrated image, its algorithm is as follows:
Micron) and the pixel count N1 that streaks 1. in target image, draw a horizontal linear section, write down starting point coordinate (x1, y) and terminal point coordinate (x2, y), and calculate the length L 1=|x1-x2|(unit of this line segment:;
Micron) and the pixel count N2 that streaks 2. in target image, draw a vertical line segment, write down starting point coordinate (x, y1) and terminal point coordinate (x, y2), and calculate the length L 2=|y1-y2|(unit of this line segment:;
3. set the enlargement factor A of this metallic phase image.
In above formula:
-two-dimensional the factor, is the two-dimentional physical size of each pixel;
(2) each crystal grain in same image is carried out to region labeling, each grained region pixel is identified, and further obtain their characteristic parameters separately.This region labeling algorithm refers to another patent of invention of the inventor: this region labeling algorithm refers to another patent of invention of the inventor: (" the automatic measurement of precipitation particles and typoiogical classification method thereof in a kind of steel ", application number: 200910030216.1).This region labeling algorithm is recursion marking algorithm, the steps include:
1. first press from left to right, the mode with TV grating starts scanning from the upper left corner of image from top to bottom.Until find one, do not have markd 1 pixel.
2. to this 1 pixel, give a new mark NewFlag.
3. press the numeral order of figure, 8 adjoint points of this object pixel (shade) point are scanned, if run into, do not have markd 1 pixel just it to be labeled as to NewFlag (it is the NewFlag in 2. namely).Now, again by 8 adjoint points of 1 pixel in 8 adjoint points of above-mentioned order scanning, as run into, do not have markd 1 pixel, again it is labeled as to NewFlag.This process is a recurrence, runs into and do not have markd 1 pixel in adjoint point, and recursion one deck, until do not have markd 1 pixel depleted, just starts to return, and returning is also to return layer by layer.
4. recurrence finishes, and continues scanning and does not have markd 1 pixel, then carries out 2., 3. two steps.
5. repeatedly carry out said process until raster scanning to the lower right corner of image.
(3) extract the geometric shape characteristic parameter of image, specific as follows:
1. particle area:
Bianry image template array is scanned, calculate the pixel sum that in target area, ash value is 255
n i , can draw target area area
a i :
2. crystal grain girth:
Adopt Freeman chain code to carry out traverse scanning to bianry image template array, the border of tracking target grained region, becomes 8 direction chain codes by frontier point coordinate conversion, and (zone boundary outline line is connected and forms piecemeal by the short line between adjacent boundary pixel.The slope of short line only may have eight directions, and 0
°, 45
°, 90
°, 135
°, 180
°, 225
°, 270
°, 315
°, with 0,1,2,3,4,5,6,7 numbers, represent respectively, be called chain code
ci=0,1,, 7}.) can draw target area crystal grain girth
i :
In formula:
n 1 -horizontal direction chain code number, 0
°, 180
°the borderline pixel sum of direction
n 2 -vertical direction chain code number, 90
°, 270
°the borderline pixel sum of direction
n 3 -oblique chain code number, 45
°, 135
°, 225
°, 315
°the borderline pixel sum of direction.
3. size of microcrystal:
Size of microcrystal
d i be: the diameter of a circle when contour area with crystal grain in image equates,
4. crystal grain circularity:
Calculate crystal grain circularity
i :
In formula:
a i -region area;
p i -area circumference
5. grain form coefficient:
Grain form coefficient
:
In formula:
;
-region area
6. crystal grain length breadth ratio:
Get the minimum boundary rectangle of target area,
w i -rectangle is wide,
l i -rectangle is long, can draw particle length breadth ratio
i:
The present invention be take size of microcrystal, as criterion, crystallite dimension is carried out to hierarchical statistics, and the circularity, shape coefficient, length breadth ratio of crystal grain of take classified to grain form as criterion.The statistic of classification result of the crystallite dimension of embodiment 1, grain form is respectively as shown in Figure 10 a, 10b.
embodiment 2
The original metallic phase image of ultra-fine grain steel as shown in figure 11, its crystal grain is tiny, different.The process that current series invention is processed it is: first target image is carried out to pre-service and the auto-thresholding algorithm divided based on region carries out binary segmentation, treatment effect as shown in figure 12; Bianry image is carried out the processing of border reparation, the filling of intracrystalline hole, treatment effect as shown in figure 13 again; Set scale and each crystal grain is carried out to region labeling, measuring and calculate the grain form characteristic parameters such as chip area, girth, length breadth ratio, diameter, circularity and shape coefficient, crystallite dimension and form are carried out to statistic of classification.The statistic of classification result of the crystallite dimension of embodiment 2, grain form is respectively as shown in Figure 14 a, 14b.
the present invention also has fabulous crystal grain measuring and classification effect, the metallographic original image of ordinary steel as shown in figure 15, its coarse grains to the common iron of a large amount of crystallite dimensions of using 20 microns of left and right in mechanical engineering.The process that current series invention is processed it is: first target image is carried out to pre-service and the auto-thresholding algorithm divided based on region carries out binary segmentation, treatment effect as shown in figure 16; Bianry image is carried out the processing of border reparation, the filling of intracrystalline hole, treatment effect as shown in figure 17 again; Set scale and each crystal grain is carried out to region labeling, measuring and calculate the grain form characteristic parameters such as chip area, girth, length breadth ratio, diameter, circularity and shape coefficient, crystallite dimension and form are carried out to statistic of classification.The statistic of classification result of the crystallite dimension of embodiment 3, grain form is respectively as shown in Figure 18 a, 18b.
Claims (3)
1. the automatic measurement of ultra-fine grain crystalline grain of steel and a typoiogical classification method thereof, is characterized in that adopting the following step:
(1) gather ultra-fine grain crystalline grain of steel image, and carry out pre-service, its concrete steps are:
(1-1) utilize the histogram equalization algorithm that can retain image detail to strengthen entire image;
(1-2) utilize rim detection method of differential operator to extract edge, the some place of gray scale sudden change is considered as to corresponding frontier point, and then the point set on definite border;
(1-3) utilize stretching algorithm to strengthen the contrast of image simultaneously;
(2) adopt the auto-thresholding algorithm of dividing based on region to carry out binary segmentation to pretreated image, obtain bianry image;
(3) described bianry image is repaired to crystal boundary by the correction watershed algorithm based on range conversion, and fill intracrystalline hole with improving seed fill algorithm, obtain repairing image; The described correction watershed algorithm based on range conversion, the steps include:
(3-1) carry out Euclidean Distance Transform, obtain each independent nucleus;
(3-2) according to correction factor, successively expand each independent nucleus, two independent nucleus adhesions after revising, are regarded as an independent nucleus, Unified number;
(3-3) nucleus after described numbering is carried out to expansion process, in expansion process, nucleus keeps increasing with layer position, when two nucleus meet, is watershed divide, now forms the separatrix of crystal grain;
(4) extract grain form characteristic parameter, its concrete steps are:
(4-1) described reparation image is carried out to scale setting and region labeling;
(4-2) extract grain form characteristic parameter: area, girth, length breadth ratio, diameter, circularity and shape coefficient;
(5) take described diameter carries out hierarchical statistics as criterion to crystallite dimension, take described circularity, shape coefficient, length breadth ratio grain form to be classified as criterion.
2. the automatic measurement of a kind of ultra-fine grain crystalline grain of steel according to claim 1 and typoiogical classification method thereof, it is characterized in that: the auto-thresholding algorithm that described step (2) is divided based on region adopts large Tianjin method algorithm, the subregion number that its region is divided is 2500.
3. the automatic measurement of a kind of ultra-fine grain crystalline grain of steel according to claim 1 and typoiogical classification method thereof, is characterized in that: described correction factor is 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110368047.XA CN102494976B (en) | 2011-11-18 | 2011-11-18 | Method for automatic measurement and morphological classification statistic of ultra-fine grain steel grains |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110368047.XA CN102494976B (en) | 2011-11-18 | 2011-11-18 | Method for automatic measurement and morphological classification statistic of ultra-fine grain steel grains |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102494976A CN102494976A (en) | 2012-06-13 |
| CN102494976B true CN102494976B (en) | 2014-04-09 |
Family
ID=46186818
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110368047.XA Expired - Fee Related CN102494976B (en) | 2011-11-18 | 2011-11-18 | Method for automatic measurement and morphological classification statistic of ultra-fine grain steel grains |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102494976B (en) |
Families Citing this family (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103018154B (en) * | 2012-12-05 | 2016-04-27 | 中国电子科技集团公司第四十五研究所 | Multi-region image detection method for traversing irregular regions of crystal grains |
| CN103886579B (en) * | 2013-12-11 | 2017-02-08 | 西安交通大学 | Abrasive particle chain self-adaptive segmentation method orienting online ferrographic image automatic identification |
| CN103903266B (en) * | 2014-04-08 | 2016-07-06 | 上海交通大学 | A kind of analyzing evaluation method of micro-nano granules dispersed and distributed |
| CN103940708B (en) * | 2014-04-10 | 2016-04-06 | 江苏大学 | A kind of Quick Measurement, sophisticated category method of steel holotype state crystal grain |
| CN104778684B (en) * | 2015-03-06 | 2019-01-08 | 江苏大学 | Automatic measurement, characterization classification method and its system of steel surface heterogeneous type defect |
| CN104729961B (en) * | 2015-03-27 | 2017-06-13 | 电子科技大学 | The detection method of particle in a kind of anisotropic conductive film |
| CN105510195B (en) * | 2015-12-07 | 2017-11-17 | 华侨大学 | A kind of granularity particle shape online test method for stacking aggregate |
| CN105527461B (en) * | 2016-01-16 | 2018-04-27 | 北京工业大学 | A kind of material structure Quantitative Analysis Method based on transmission electron microscope HAADF images |
| CN105719291A (en) * | 2016-01-20 | 2016-06-29 | 江苏省沙钢钢铁研究院有限公司 | Surface defect image classification system having selectable types |
| CN106023134B (en) * | 2016-04-29 | 2018-08-10 | 江苏大学 | A kind of automatic crystal boundary extracting method of steel crystal grain |
| CN108226159B (en) * | 2017-12-29 | 2019-11-22 | 钢铁研究总院 | The full filed quantitative statistics of precipitated phase particle are distributed characterizing method in metal material |
| CN109612888B (en) * | 2018-12-12 | 2021-06-22 | 太原理工大学 | Powder mixing uniformity detection method based on image technology |
| CN111325706A (en) * | 2018-12-13 | 2020-06-23 | 阿里巴巴集团控股有限公司 | Grain boundary extraction and grain boundary extraction model generation method and device |
| CN109615630A (en) * | 2019-01-18 | 2019-04-12 | 东北大学 | Semi-continuous casting alusil alloy Analysis on Microstructure method based on image processing techniques |
| CN109961451A (en) * | 2019-03-22 | 2019-07-02 | 西北工业大学 | A kind of material grains tissue segmentation methods based on marginal information |
| CN110595959B (en) * | 2019-04-19 | 2023-03-21 | 齐鲁工业大学 | Massecuite particle size analysis system and analysis method |
| CN110057650B (en) * | 2019-05-20 | 2022-03-15 | 常德力元新材料有限责任公司 | Method for evaluating steel strip crystal grains |
| CN110361385A (en) * | 2019-07-08 | 2019-10-22 | 武汉科技大学 | A kind of method and system of crystal grain data acquisition |
| CN112945765A (en) * | 2019-12-11 | 2021-06-11 | 宝武特种冶金有限公司 | Titanium alloy reversing forging test method based on numerical simulation |
| CN113125319A (en) * | 2021-05-25 | 2021-07-16 | 西安热工研究院有限公司 | Method for measuring and characterizing grain size of non-equiaxed grain structure of austenitic stainless steel pipe |
| CN115861326B (en) * | 2023-03-01 | 2023-05-16 | 山东神力索具有限公司 | Image processing-based steel grain size detection method for lifting hook |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101510262A (en) * | 2009-03-17 | 2009-08-19 | 江苏大学 | Automatic measurement method for separated-out particles in steel and morphology classification method thereof |
| CN101964293A (en) * | 2010-08-23 | 2011-02-02 | 西安航空动力股份有限公司 | Metallographical microstructural image processing method |
| CN102222349A (en) * | 2011-07-04 | 2011-10-19 | 江苏大学 | Prospect frame detecting method based on edge model |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5307695B2 (en) * | 2009-11-16 | 2013-10-02 | 大成建設株式会社 | Granular material particle size identification method and particle size identification system |
-
2011
- 2011-11-18 CN CN201110368047.XA patent/CN102494976B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101510262A (en) * | 2009-03-17 | 2009-08-19 | 江苏大学 | Automatic measurement method for separated-out particles in steel and morphology classification method thereof |
| CN101964293A (en) * | 2010-08-23 | 2011-02-02 | 西安航空动力股份有限公司 | Metallographical microstructural image processing method |
| CN102222349A (en) * | 2011-07-04 | 2011-10-19 | 江苏大学 | Prospect frame detecting method based on edge model |
Non-Patent Citations (1)
| Title |
|---|
| JP特开2011-106923A 2011.06.02 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102494976A (en) | 2012-06-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102494976B (en) | Method for automatic measurement and morphological classification statistic of ultra-fine grain steel grains | |
| CN111145161B (en) | Pavement crack digital image processing and identifying method | |
| CN107808378B (en) | Method for detecting potential defects of complex-structure casting based on vertical longitudinal and transverse line profile features | |
| CN112419250B (en) | Pavement crack digital image extraction, crack repair and crack parameter calculation method | |
| CN109978839B (en) | Method for detecting wafer low-texture defects | |
| CN108629775B (en) | Thermal state high-speed wire rod surface image processing method | |
| CN109580630B (en) | Visual inspection method for defects of mechanical parts | |
| CN109658381B (en) | Method for detecting copper surface defects of flexible IC packaging substrate based on super-pixels | |
| CN107369136B (en) | Visual detection method for surface cracks of polycrystalline diamond compact | |
| CN102496161B (en) | Method for extracting contour of image of printed circuit board (PCB) | |
| CN103940708B (en) | A kind of Quick Measurement, sophisticated category method of steel holotype state crystal grain | |
| CN104880389A (en) | Mixed crystal degree automatic measurement and fine classification method for steel crystal grains, and system thereof | |
| CN104778684B (en) | Automatic measurement, characterization classification method and its system of steel surface heterogeneous type defect | |
| CN102054168A (en) | Method for recognizing circular seal on valuable bill | |
| CN110287898A (en) | A kind of optical satellite remote sensing image cloud detection method of optic | |
| CN111354047B (en) | Computer vision-based camera module positioning method and system | |
| CN105741281B (en) | Method for detecting image edge based on neighborhood dispersion | |
| CN105427324A (en) | Magneto-optic imaging defect detection method based on automatic search of binary threshold | |
| CN110567965A (en) | Smartphone glass cover plate edge visual defect detection method | |
| CN114926407A (en) | Steel surface defect detection system based on deep learning | |
| CN112734761A (en) | Industrial product image boundary contour extraction method | |
| CN105787950A (en) | Infrared image sea-sky-line detection algorithm based on line gradient accumulation | |
| CN105678763B (en) | A kind of stable straight line/circle characteristic detection method and device | |
| CN108492306A (en) | A kind of X-type Angular Point Extracting Method based on image outline | |
| CN109064420B (en) | Multilayer PCB section image process parameter extraction method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20140409 Termination date: 20151118 |
|
| EXPY | Termination of patent right or utility model |