CN110853017A - Method for counting number of round particles in powder particle image - Google Patents
Method for counting number of round particles in powder particle image Download PDFInfo
- Publication number
- CN110853017A CN110853017A CN201911104266.XA CN201911104266A CN110853017A CN 110853017 A CN110853017 A CN 110853017A CN 201911104266 A CN201911104266 A CN 201911104266A CN 110853017 A CN110853017 A CN 110853017A
- Authority
- CN
- China
- Prior art keywords
- target
- circle
- circles
- counting
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
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
- G06T7/11—Region-based segmentation
-
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10056—Microscopic image
-
- 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
-
- 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
- G06T2207/30242—Counting objects in image
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 discloses a method for counting the number of round particles in a powder particle image, which comprises the following steps: extracting all communication domains in the image, and judging that each communication domain is a single round target or an overlapped round target; for the overlapping circle target, the diameter and number of circles are counted using a center drift algorithm. The invention has high algorithm precision and is used for counting the quantity and the size of the powder particles with irregular shapes, different sizes and overlapping conditions in the powder image.
Description
Technical Field
The invention relates to a powder particle quality detection technology, in particular to a method for counting round particles in a powder particle image.
Background
Ultrafine powder particles are widely applied to the industrial fields of materials, manufacturing and the like. The quantity and the size of the powder particles in a unit area are important indexes for measuring the quality of the powder, and the method has great application value in the aspects of additive manufacturing and the like by utilizing metal powder, however, no effective method for counting ultrafine powder particles exists at present, and a simple statistical method is integrated in some crystal image microscopes, but the statistical accuracy is low, effective statistics cannot be carried out on overlapped particle areas, and the evaluation result of the quality of the powder particles is seriously influenced.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a method for counting the number and the size of powder particles under the conditions of irregular shape, different size and overlapping in a powder image, and counting the number of round particles in the powder particle image with high algorithm precision.
The purpose of the invention is realized by the following technical scheme.
A method for counting the number of round particles in a powder particle image comprises the following steps:
1) extracting all communication domains in the image, and judging that each communication domain is a single round target or an overlapped round target;
2) for the overlapping circle target, the diameter and number of circles are counted using a center drift algorithm.
The step 1) is specifically as follows:
1-1) aiming at the whole image, extracting the powder particle foreground by using a threshold segmentation method;
1-2) converting the foreground image into a binary image, and marking 8 communication domains in the image to obtain N targets;
1-3) solving the minimum circumcircle of the contour aiming at each target, and recording the radius Rn of the minimum circumcircle;
1-4) aiming at each target, calculating the ratio y of the target area S1 to the circumscribed circle area S2, if y is greater than 0.85, the target is a single circle target, adding 1 to the number of the circle targets, and taking the radius of the circumscribed circle as the radius of the circle; otherwise, the target will be considered as an overlapping circle target.
The step 2) is specifically as follows:
2-1) extracting a target edge for each overlapped target;
2-2) randomly selecting points in the target, the number of which is in direct proportion to the area of the target, as the circle center;
2-3) constructing a circle with the radius of 1 at each circle center;
2-4) gradually increasing the radius of each circle, when the circles are intersected with the edge of the target, moving the circle center towards the opposite direction of the intersection point, gradually increasing the radii of a plurality of circles and enabling the circles to collide back and forth in the target, combining the two circles with the distance between the circle centers smaller than a threshold value L into a circle, and finally enabling the circle in the target to be static;
2-5) counting the final distribution situation of the target middle circles, wherein the number of the target middle circles is the number of the overlapped target middle circles, and counting the radius of each circle;
2-6) counting the number of circles in the whole image and the radius of each circle and classifying to obtain the counting result of the number of the circles.
Compared with the prior art, the invention has the advantages that: the invention provides a method for accurately counting the number of ultrafine powder particles, and can be widely applied to the fields of ultrafine powder characteristic analysis, metal powder quality identification in additive manufacturing and the like. The invention can easily judge whether the target is a single target or an overlapped target, can count the number of overlapped particles, has high algorithm progress which can reach 98 percent and strong robustness, and can greatly improve the counting and analyzing quality of the number of the particles compared with the prior art.
Drawings
FIG. 1 is a schematic diagram of the diameter and number of overlapping target statistic circles of the present invention.
Fig. 2 is a schematic diagram of an image to be extracted.
Detailed Description
The invention is described in detail below with reference to the drawings and specific examples.
A method for counting the number of round particles in a powder particle image comprises the following steps:
1) extracting all communication domains in the image shown in FIG. 2, and judging that each communication domain is a single round target or an overlapped round target;
2) for the overlapping circle target, the diameter and number of circles are counted using a center drift algorithm.
The step 1) is specifically as follows:
1-1) extracting powder particle foreground by using a threshold segmentation method aiming at the whole image, wherein the threshold T is 120;
1-2) converting the foreground image into a binary image, and marking 8 communication domains in the image to obtain N targets;
1-3) solving the minimum circumcircle of the contour aiming at each target, and recording the radius Rn of the minimum circumcircle;
1-4) aiming at each target, calculating the ratio y of the target area S1 to the circumscribed circle area S2, if y is greater than 0.85, the target is a single circle target, adding 1 to the number of the circle targets, and taking the radius of the circumscribed circle as the radius of the circle; otherwise, the target will be considered as an overlapping circle target.
The step 2) is specifically as follows:
2-1) for each overlapped object, as shown in fig. 1(a), there are objects (two circles) of the overlapped area, extracting the object edge, as shown in fig. 1 (b);
2-2) randomly selecting points in the target, the number of which is in direct proportion to the area of the target, as circle centers, wherein the number of the circle centers is the area of the target, and performing modulo operation on the area of the target and the area of the circle centers by using 80, namely S mod (80);
2-3) constructing a circle with the radius of 1 at each circle center, as shown in fig. 1(c), randomly scattering a plurality of seed points (three points) in the target, and generating a corresponding number of circles, wherein the radius starts from 1 and increases from small to large;
2-4) gradually increasing the radius of each circle, when the circles intersect with the edge of the target, the circle center moves towards the opposite direction of the intersection point, so that the radiuses of a plurality of circles gradually increase and collide back and forth in the target, as shown in fig. 1(d), the circles "bounce" when contacting the inside of the edge of the target in the gradually increasing process, and the distance between the circle centers is smaller than a threshold value L (the threshold value is 0.2 times the radius of the circumscribed circle, namely L is 0.2Rn) The two circles are merged into one circle, and finally the circle in the target is static, as shown in fig. 1(e), finally, the positions and the radiuses of a plurality of circles are not changed any more, and the two circles in the target can be known by counting the position of the final circle center;
2-5) counting the final distribution situation of the target middle circles, wherein the number of the target middle circles is the number of the overlapped target middle circles, and counting the radius of each circle;
2-6) counting the number of circles in the whole image and the radius of each circle and classifying to obtain the counting result of the number of the circles.
Serial number | Image name | 1 to 10 pixels | 11 to 20 pixels | 21 to 30 pixels | 31 to 40 pixels | 41 to 50 pixels |
1 | XBCD.jpg | 174 | 389 | 1162 | 389 | 0 |
Claims (3)
1. A method for counting the number of round particles in a powder particle image is characterized by comprising the following steps:
1) extracting all communication domains in the image, and judging that each communication domain is a single round target or an overlapped round target;
2) for the overlapping circle target, the diameter and number of circles are counted using a center drift algorithm.
2. The method for counting the number of round particles in a powder particle image according to claim 1, wherein the step 1) is specifically as follows:
1-1) aiming at the whole image, extracting the powder particle foreground by using a threshold segmentation method;
1-2) converting the foreground image into a binary image, and marking 8 communication domains in the image to obtain N targets;
1-3) solving the minimum circumcircle of the contour aiming at each target, and recording the radius Rn of the minimum circumcircle;
1-4) aiming at each target, calculating the ratio y of the target area S1 to the circumscribed circle area S2, if y is greater than 0.85, the target is a single circle target, adding 1 to the number of the circle targets, and taking the radius of the circumscribed circle as the radius of the circle; otherwise, the target will be considered as an overlapping circle target.
3. The method for counting the number of round particles in the powder particle image according to claim 1 or 2, wherein the step 2) is specifically as follows:
2-1) extracting a target edge for each overlapped target;
2-2) randomly selecting points in the target, the number of which is in direct proportion to the area of the target, as the circle center;
2-3) constructing a circle with the radius of 1 at each circle center;
2-4) gradually increasing the radius of each circle, when the circles intersect with the edge of the target, the circle center moves towards the opposite direction of the intersection point, the radiuses of a plurality of circles are gradually increased and collide back and forth in the target, and the distance between the circle centers is smaller than a threshold value L, namely L is 0.2RnThe two circles are merged into one circle, and the circle in the final target is static;
2-5) counting the final distribution situation of the target middle circles, wherein the number of the target middle circles is the number of the overlapped target middle circles, and counting the radius of each circle;
2-6) counting the number of circles in the whole image and the radius of each circle and classifying to obtain the counting result of the number of the circles.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911104266.XA CN110853017A (en) | 2019-11-13 | 2019-11-13 | Method for counting number of round particles in powder particle image |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911104266.XA CN110853017A (en) | 2019-11-13 | 2019-11-13 | Method for counting number of round particles in powder particle image |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110853017A true CN110853017A (en) | 2020-02-28 |
Family
ID=69601062
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911104266.XA Withdrawn CN110853017A (en) | 2019-11-13 | 2019-11-13 | Method for counting number of round particles in powder particle image |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110853017A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112504947A (en) * | 2020-12-03 | 2021-03-16 | 中国人民解放军陆军军医大学第二附属医院 | Morphological analysis and counting method for peripheral blood cells |
-
2019
- 2019-11-13 CN CN201911104266.XA patent/CN110853017A/en not_active Withdrawn
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112504947A (en) * | 2020-12-03 | 2021-03-16 | 中国人民解放军陆军军医大学第二附属医院 | Morphological analysis and counting method for peripheral blood cells |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108376403B (en) | Grid colony image segmentation method based on Hough circle transformation | |
CN103593695B (en) | A kind of method of DPM two-dimension code area location | |
CN103295227B (en) | A kind of circle detection method based on gradient direction segmentation | |
CN114926463A (en) | Production quality detection method suitable for chip circuit board | |
Wu et al. | Watershed-based morphological separation of wear debris chains for on-line ferrograph analysis | |
CN109523565A (en) | A kind of diffraction light-free Moire fringe center positioning method and system | |
CN112986925B (en) | Radar pulse sequence sorting method based on image features | |
CN115249246A (en) | Optical glass surface defect detection method | |
Wälder et al. | On variograms in point process statistics | |
CN106251332A (en) | SAR image airport target detection method based on edge feature | |
CN110895333A (en) | Rapid 77G vehicle-mounted radar data clustering method based on Doppler frequency | |
CN111553886B (en) | Self-adaptive identification method for particles in CT image of asphalt mixture | |
CN116309577B (en) | Intelligent detection method and system for high-strength conveyor belt materials | |
CN110853017A (en) | Method for counting number of round particles in powder particle image | |
CN115965624A (en) | Detection method for anti-wear hydraulic oil pollution particles | |
CN103077499A (en) | SAR (Synthetic Aperture Radar) image pre-processing method based on similar block | |
CN104648397A (en) | System and method for warning lane departure | |
CN109271997B (en) | Image texture classification method based on skip subdivision local mode | |
US20150243008A1 (en) | Pattern Measurement Device, Evaluation Method of Polymer Compounds Used in Self-Assembly Lithography, and Computer Program | |
Andersson et al. | Minimizing profile error when estimating the sieve-size distribution of iron ore pellets using ordinal logistic regression | |
CN113191330A (en) | Method for identifying area growth pores by fusing secondary electrons and back scattering electron images | |
CN107967690B (en) | Self-adaptive ferrographic abrasive particle image binarization processing method | |
CN116540185A (en) | Sorting identification method based on density clustering | |
CN105069403A (en) | Three-dimensional ear recognition based on block statistic features and dictionary learning sparse representation classification | |
CN115272946B (en) | Method for identifying damage of common rail fuel injector by using electronic equipment |
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 | ||
WW01 | Invention patent application withdrawn after publication | ||
WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200228 |