CN113538345A - Industrial bottle and can device counting method based on image processing - Google Patents
Industrial bottle and can device counting method based on image processing Download PDFInfo
- Publication number
- CN113538345A CN113538345A CN202110726962.5A CN202110726962A CN113538345A CN 113538345 A CN113538345 A CN 113538345A CN 202110726962 A CN202110726962 A CN 202110726962A CN 113538345 A CN113538345 A CN 113538345A
- Authority
- CN
- China
- Prior art keywords
- image
- bottle
- industrial
- counting method
- devices
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000001914 filtration Methods 0.000 claims abstract description 13
- 230000011218 segmentation Effects 0.000 claims abstract description 8
- 230000000877 morphologic effect Effects 0.000 claims abstract description 7
- 238000005260 corrosion Methods 0.000 claims abstract description 4
- 230000007797 corrosion Effects 0.000 claims abstract description 4
- 238000005516 engineering process Methods 0.000 claims abstract description 4
- 230000009466 transformation Effects 0.000 claims description 9
- 230000003628 erosive effect Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 5
- 230000010339 dilation Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 abstract description 3
- 238000009776 industrial production Methods 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 2
- 208000003464 asthenopia Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 235000013361 beverage Nutrition 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 235000013365 dairy product Nutrition 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
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
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration by the use of local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration by the use of histogram techniques
-
- G06T5/70—
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20036—Morphological image processing
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20112—Image segmentation details
- G06T2207/20152—Watershed segmentation
-
- 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)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
An industrial bottle and can device counting method based on image processing is characterized in that an industrial camera takes a snapshot of an industrial bottle and can device picture as a data source and transmits the snapshot to a computer image preprocessing part; carrying out image enhancement on the original industrial bottle device image by a histogram equalization enhancement technology; then, a wiener filtering algorithm is adopted to reduce noise of the image, and meanwhile, the edge profile of the bottle device is kept well; after image preprocessing is carried out, the bottle device and the background are segmented by adopting an OTSU maximum between-class variance algorithm to obtain an ideal bottle device binary image; removing false from the binary image of the bottle device by morphological expansion and corrosion operation; separating the adhered bottle and can devices by using a watershed segmentation algorithm; finally, the number of devices is counted based on the size area threshold. The device and the method are suitable for counting the number of the industrial bottle and can devices, replace the traditional manual counting mode, save the labor cost and improve the working efficiency of industrial production.
Description
Technical Field
The invention relates to the field of image processing, in particular to an industrial bottle and can device counting method based on image processing.
Background
The production of bottle-can type devices is huge in industry, and the bottle-can type devices are widely used, such as beverage bottles, dairy product packages, liquid medical medicine containers and the like. The traditional counting method for devices by manufacturers is usually a manual counting method, the counting method consumes a large amount of manpower time and is low in efficiency, and eye fatigue is easily caused due to long-time checking and counting by naked eyes, so that result errors are large.
Disclosure of Invention
In order to solve the problems in the background art, the invention provides the image processing-based industrial bottle and can device counting method which is simple and easy to operate, has accurate and efficient calculation, can effectively count industrial bottle and can devices, replaces a manual counting mode, saves the labor cost and greatly improves the industrial work efficiency.
The technical scheme adopted by the invention for solving the technical problems is as follows:
an industrial bottle and can device counting method based on image processing comprises the following steps:
step S1: overlooking and shooting images of the bottle and can devices, and transmitting the images to a computer system as a data source;
step S2: carrying out image enhancement on the original industrial bottle device image by a histogram equalization enhancement technology;
step S3: carrying out noise reduction processing on the original image by adopting a wiener filtering algorithm;
step S4: segmenting the bottle device and the background by adopting an OTSU maximum between-class variance algorithm to obtain an ideal bottle device binary image;
step S5: removing false operation of the binary image of the bottle device by adopting morphological expansion and corrosion operation;
step S6: separating the adhered bottle and can devices by using a watershed segmentation algorithm;
step S7: finally, the number of devices is counted based on the size area threshold.
Further, in step S2, the histogram is a visual statistic of the gray level distribution range of the image, and the gray level histogram function represents the gray level range of [0, L-1 ].
h(rp)=np
Where np is the number of pixels whose image gray scale is rp, and rp is the p-th gray scale in the image.
Still further, in step S3, the wiener filter is an optimal linear filtering method proposed according to the minimum mean square error criterion, and the application range is wide. When the image is subjected to wiener filtering, the mean and variance of the local matrix of the target pixel a (n1, n2) are estimated, and the calculation formula is as follows:
wherein η is a neighborhood of M × N of a certain pixel;
then estimating the transformed pixel value b (n1, n2) by using a wiener filter;
wherein v is2Is the image variance.
In step S4, the image is binarized by using the maximum inter-class variance algorithm of the OSTU, and a threshold T is obtained.
In step S5, morphological operations such as erosion and dilation are performed on the extracted binary image, so that noise can be effectively eliminated, and a complete device region and clear edge information can be obtained.
In step S6, a distance-based watershed segmentation method is used, and after distance change is performed on the binary image, a distance set from each pixel in the binary image to a pixel closest to a zero value is obtained, and watershed transformation is performed according to the set;
complementing the target binary image to obtain a distance transformation function:
D(p)=min{dist(p,q),q∈m}
the region of interest is m, D represents a set after distance transformation, p and q are pixel points, and dist is a distance function.
After the processing of steps S2, S3, S4, S5, and S6, the target regions are basically binary images of the bottle-can devices in a plan view state separated from each other, and a counting method based on a size area is adopted for the statistical number.
According to the method, firstly, the image of the device is enhanced through a histogram equalization enhancement technology, then the original image is subjected to primary noise reduction processing through wiener filtering, after a binary image of a more ideal original image is extracted through a maximum between-class variance algorithm, false removing processing is performed through morphological corrosion and expansion operation, the secondary noise reduction effect is achieved, bottle and can devices in the binary image are mutually independent through a better segmentation effect of a watershed algorithm, and therefore counting is achieved through a counting method based on the size area.
The invention has the following beneficial effects: the counting machine is simple and easy to operate, is accurate and efficient in calculation, can effectively count industrial bottle devices, replaces a manual counting mode, saves labor cost and greatly improves industrial work efficiency.
Drawings
FIG. 1 is a system flow diagram of an industrial can device counting method based on image processing.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
Referring to fig. 1, an image processing-based industrial bottle and can device counting method includes the following steps:
step S1: image acquisition
The high-definition images of the bottle and can devices can be obtained by a high-definition camera in a way of overhead shooting, the images are clear, have no shielding and are not obvious, the shot images are used as data sources of the method, the number of the bottle and can devices is counted, the arrangement modes of the same kind of bottle and can devices in the images are random, and then the images are transmitted to a computer for processing and counting.
Step 2: histogram equalization enhancement
Preprocessing a target image, performing image enhancement on the image of the bottle device by utilizing a gray level histogram, enhancing the contrast ratio of a device region and a background, and providing a more real digital image for noise reduction, false removal and counting of a subsequent image;
the histogram is a visual statistic of the gray level distribution range of the image, with the gray level histogram function representing the gray level range of [0, L-1 ].
h(rp)=np
Np is the number of pixels with the image gray level rp, and rp is the p-th gray level in the image;
and step 3: wiener filtering noise reduction
Carrying out first noise reduction on the original picture subjected to histogram equalization enhancement through wiener filtering, wherein the noise reduction is carried out aiming at eliminating noise caused by factors such as shooting angle, light intensity and vibration and simultaneously keeping the edge outline of a bottle device in the image;
wiener filtering is an optimal linear filtering method provided according to the minimum mean square error criterion, and has a wide application range. When the image is subjected to wiener filtering, the mean and variance of the local matrix of the target pixel a (n1, n2) are estimated, and the calculation formula is as follows:
wherein η is a neighborhood of M × N of a certain pixel;
then estimating the transformed pixel value b (n1, n2) by using a wiener filter;
wherein v is2Is the image variance.
And 4, step 4: OTSU binary image extraction
After image processing, what is needed for counting is a binary image of the can device. The OTSU maximum between-class variance algorithm is based on a clustering idea, data is divided into two categories by using a threshold T, when the variance between the thresholds is maximum, the probability of wrong division is minimum, so that the gray value of a pixel point with the gray value larger than the threshold T is adjusted to be 255 again, and the gray value of a pixel point with the gray value smaller than the threshold T is adjusted to be 0. The device and the background are segmented by the algorithm, and an ideal device binary image is obtained.
And 5: morphological operations to remove artifacts
Erosion and dilation are basic operations of mathematical morphology, where erosion reduces the area of a white region, while dilation is the expansion of a white region in an image, and the pixel area is changed by erosion and erosion to obtain an ideal output image.
Step 6: watershed segmentation based on distance
The watershed algorithm based on distance can effectively separate the adhered objects. For the condition that a plurality of bottle devices are adhered and abutted in the image, the device is relatively independent by complementing the target image to obtain distance conversion and then segmenting, so that a subsequent counting method is conveniently utilized;
adopting a distance-based watershed segmentation method, obtaining a distance set from each pixel in the binary image to a pixel closest to a zero value after distance change is carried out on the binary image, and carrying out watershed transformation according to the set;
complementing the target binary image to obtain a distance transformation function:
D(p)=min{dist(p,q),q∈m}
the region of interest is m, D represents a set after distance transformation, p and q are pixel points, and dist is a distance function.
And 7: counting method based on size area
After the processing of steps S2, S3, S4, S5 and S6, the target regions are basically binary images of the bottle-can devices in a mutually separated, overlooked state, and a counting method based on size and area is adopted for counting the number;
after the images are processed for a plurality of times in the previous period, the target areas are basically separated from each other, 2/3 sizes of corresponding actual bottle size areas are selected as judgment threshold values S in the algorithm process, when the areas are larger than the threshold values S, the devices are determined, the number is increased by 1, otherwise, no operation is performed, and finally the number of the devices in the input bottle device images can be obtained.
The embodiments described in this specification are merely illustrative of implementations of the inventive concepts, which are intended for purposes of illustration only. The scope of the present invention should not be construed as being limited to the particular forms set forth in the examples, but rather as being defined by the claims and the equivalents thereof which can occur to those skilled in the art upon consideration of the present inventive concept.
Claims (7)
1. An industrial bottle and can device counting method based on image processing is characterized by comprising the following steps:
step S1: overlooking and shooting images of the bottle and can devices, and transmitting the images to a computer system as a data source;
step S2: carrying out image enhancement on the original industrial bottle device image by a histogram equalization enhancement technology;
step S3: carrying out noise reduction processing on the original image by adopting a wiener filtering algorithm;
step S4: segmenting the bottle device and the background by adopting an OTSU maximum between-class variance algorithm to obtain an ideal bottle device binary image;
step S5: removing false operation of the binary image of the bottle device by adopting morphological expansion and corrosion operation;
step S6: separating the adhered bottle and can devices by using a watershed segmentation algorithm;
step S7: finally, the number of devices is counted based on the size area threshold.
2. The image processing based industrial canister device counting method according to claim 1, characterized in that: in step S2, the histogram is a visual statistic of the gray-level distribution range of the image, and the gray-level histogram function represents the gray-level range of [0, L-1 ].
h(rp)=np
Where np is the number of pixels whose image gray scale is rp, and rp is the p-th gray scale in the image.
3. Image processing based industrial bottle and can device counting method according to claim 1 or 2, characterized in that: in step S3, the wiener filter is an optimal linear filtering method proposed according to the minimum mean square error criterion, and the application range is wide. When the image is subjected to wiener filtering, the mean and variance of the local matrix of the target pixel a (n1, n2) are estimated, and the calculation formula is as follows:
μ=1/MN∑n1,n2∈ηa(n1,n2)
σ2=1/MN∑n1,n2∈ηa2(n1,n2)-μ2
wherein η is a neighborhood of M × N of a certain pixel;
then estimating the transformed pixel value b (n1, n2) by using a wiener filter;
wherein v is2Is the image variance.
4. Image processing based industrial bottle and can device counting method according to claim 1 or 2, characterized in that: in step S4, the image is binarized by using the maximum inter-class variance algorithm of the OSTU, and a threshold T is obtained.
5. Image processing based industrial bottle and can device counting method according to claim 1 or 2, characterized in that: in step S5, morphological operations such as erosion and dilation are performed on the extracted binary image, so that noise can be effectively eliminated, and a complete device region and clear edge information can be obtained.
6. Image processing based industrial bottle and can device counting method according to claim 1 or 2, characterized in that: in step S6, a distance-based watershed segmentation method is used to obtain a distance set from each pixel in the binary image to a pixel closest to a zero value after distance change is performed on the binary image, and watershed transformation is performed according to the set;
complementing the target binary image to obtain a distance transformation function:
D(p)=min{dist(p,q),q∈m}
the region of interest is m, D represents a set after distance transformation, p and q are pixel points, and dist is a distance function.
7. The image processing based industrial canister device counting method according to claim 1, characterized in that: after the processing of steps S2, S3, S4, S5, and S6, the target regions are basically binary images of the bottle-can devices in a plan view state separated from each other, and a counting method based on a size area is adopted for the statistical number.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110726962.5A CN113538345A (en) | 2021-06-29 | 2021-06-29 | Industrial bottle and can device counting method based on image processing |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110726962.5A CN113538345A (en) | 2021-06-29 | 2021-06-29 | Industrial bottle and can device counting method based on image processing |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113538345A true CN113538345A (en) | 2021-10-22 |
Family
ID=78126186
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110726962.5A Pending CN113538345A (en) | 2021-06-29 | 2021-06-29 | Industrial bottle and can device counting method based on image processing |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113538345A (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103914843A (en) * | 2014-04-04 | 2014-07-09 | 上海交通大学 | Image segmentation method based on watershed algorithm and morphological marker |
WO2016091016A1 (en) * | 2014-12-12 | 2016-06-16 | 山东大学 | Nucleus marker watershed transformation-based method for splitting adhered white blood cells |
-
2021
- 2021-06-29 CN CN202110726962.5A patent/CN113538345A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103914843A (en) * | 2014-04-04 | 2014-07-09 | 上海交通大学 | Image segmentation method based on watershed algorithm and morphological marker |
WO2016091016A1 (en) * | 2014-12-12 | 2016-06-16 | 山东大学 | Nucleus marker watershed transformation-based method for splitting adhered white blood cells |
Non-Patent Citations (2)
Title |
---|
王念富: "基于机器视觉的羊群图像分割及计数算法研究", 《中国优秀硕士学位论文全文数据库 农业科技辑》, no. 2 * |
谢勤岚;于小卉;: "一种基于Matlab的血红细胞计数的工程方法", 中南民族大学学报(自然科学版), no. 04 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108898610B (en) | Object contour extraction method based on mask-RCNN | |
CN108376403B (en) | Grid colony image segmentation method based on Hough circle transformation | |
CN110287963B (en) | OCR recognition method for comprehensive performance test | |
CN114972326A (en) | Defective product identification method for heat-shrinkable tube expanding process | |
CN110210477B (en) | Digital instrument reading identification method | |
CN110674812B (en) | Civil license plate positioning and character segmentation method facing complex background | |
CN105447489B (en) | A kind of character of picture OCR identifying system and background adhesion noise cancellation method | |
CN110648330B (en) | Defect detection method for camera glass | |
CN115063430B (en) | Electric pipeline crack detection method based on image processing | |
CN110415208A (en) | A kind of adaptive targets detection method and its device, equipment, storage medium | |
CN111754538B (en) | Threshold segmentation method for USB surface defect detection | |
CN113252695B (en) | Plastic packaging film defect detection method and detection device based on image processing | |
CN111476804A (en) | Method, device and equipment for efficiently segmenting carrier roller image and storage medium | |
CN116188468A (en) | HDMI cable transmission letter sorting intelligent control system | |
CN108038482A (en) | A kind of automobile engine cylinder-body sequence number Visual intelligent identifying system | |
CN113487538A (en) | Multi-target segmentation defect detection method and device and computer storage medium thereof | |
Al-Mahadeen et al. | Signature region of interest using auto cropping | |
CN111242051B (en) | Vehicle identification optimization method, device and storage medium | |
Wang et al. | Adaptive binarization: A new approach to license plate characters segmentation | |
CN113538345A (en) | Industrial bottle and can device counting method based on image processing | |
CN108009459B (en) | Character two-dimensional bar code rapid positioning method based on triangular locator | |
CN113643290B (en) | Straw counting method and device based on image processing and storage medium | |
CN115587966A (en) | Method and system for detecting whether parts are missing or not under condition of uneven illumination | |
CN115631128A (en) | Circular medicine quantity detection method based on circular power theorem | |
CN114758125A (en) | Gear surface defect detection method and system based on deep learning |
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 |