CN109377485B - Machine vision detection method for instant noodle packaging defects - Google Patents
Machine vision detection method for instant noodle packaging defects Download PDFInfo
- Publication number
- CN109377485B CN109377485B CN201811188773.1A CN201811188773A CN109377485B CN 109377485 B CN109377485 B CN 109377485B CN 201811188773 A CN201811188773 A CN 201811188773A CN 109377485 B CN109377485 B CN 109377485B
- Authority
- CN
- China
- Prior art keywords
- image
- gray
- packaging
- defect
- value
- 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
- 235000008446 instant noodles Nutrition 0.000 title claims abstract description 52
- 230000007547 defect Effects 0.000 title claims abstract description 49
- 238000001514 detection method Methods 0.000 title claims abstract description 48
- 238000004806 packaging method and process Methods 0.000 title claims abstract description 44
- 238000000034 method Methods 0.000 claims abstract description 17
- 238000007781 pre-processing Methods 0.000 claims abstract description 5
- 241001270131 Agaricus moelleri Species 0.000 claims abstract description 4
- 230000002708 enhancing effect Effects 0.000 claims abstract description 4
- 238000005259 measurement Methods 0.000 claims abstract description 4
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000007689 inspection Methods 0.000 claims description 5
- 238000013507 mapping Methods 0.000 claims description 5
- 241001632422 Radiola linoides Species 0.000 claims description 2
- 238000010606 normalization Methods 0.000 claims description 2
- 238000012856 packing Methods 0.000 claims 1
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 235000013305 food Nutrition 0.000 description 1
- 230000000007 visual effect Effects 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
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- 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
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- 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
- G06T2207/10024—Color 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
- G06T2207/20—Special algorithmic details
- G06T2207/20076—Probabilistic 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30128—Food products
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a machine vision detection method for defects of instant noodle packaging, which mainly comprises the following steps: establishing an automatic detection device model for the packaging defects of the instant noodles, collecting an image of the packaging of the instant noodles on a conveyor belt, preprocessing the image, and enhancing the image characteristics; selecting image segmentation center pixel points, establishing a similarity measurement standard according to the continuity between pixel grayscales, and segmenting all pixels into different regions; convolving the packaging image by a Gaussian kernel function to obtain a Gaussian difference image, and extracting the defect characteristics of the packaging image of the instant noodle by solving the gray extreme value of the image; classifying by measuring the similarity between the defect characteristics and the sample, detecting the defect category and kicking off, and completing the automatic detection of the defects of the instant noodle packaging. The method has the advantages of good stability and robustness, low omission factor and false detection rate, high detection efficiency, high detection speed, nondestructive detection, labor resource saving, cost reduction for enterprises and more fine instant noodle packaging.
Description
Technical Field
The invention relates to the fields of food quality inspection, image recognition and mathematics, in particular to a machine vision detection method for defects of instant noodle packaging.
Background
The packaging of instant noodles has a great influence on the appearance and quality of instant noodles, and various defects or flaws inevitably exist in the printing production. The existing instant noodle packaging defect detection device has the advantages of high labor intensity, low detection efficiency, poor adaptability, high detection cost and low competitiveness of instant noodles, and is easy to cause false detection and missing detection, thus being not beneficial to reducing the production cost. Enterprises without online detection devices can only manually check the defects, and the defect detection rate is very low.
Disclosure of Invention
In order to solve the problems, the invention aims to provide a machine vision detection method for defects of instant noodle packages, which has better stability and robustness, reduces the omission factor and the false detection rate, improves the detection efficiency, has high detection speed, realizes nondestructive detection, saves labor resources, reduces the cost for enterprises, and makes the instant noodle packages more refined.
The technical scheme adopted by the invention for solving the problems comprises the following steps:
A. establishing an automatic detection device model for the packaging defects of the instant noodles, collecting an image of the packaging of the instant noodles on a conveyor belt, preprocessing the image, and enhancing the image characteristics;
B. selecting an image segmentation center pixel point, establishing a similarity measurement standard according to the continuity between pixel grayscales, and segmenting all pixels into different regions;
C. convolving the segmented packaging image by a Gaussian kernel function to obtain a Gaussian difference image, and extracting the defect characteristics of the packaging image of the instant noodle by solving the gray extreme value of the image;
D. classifying by measuring the similarity between the defect characteristics and the sample, detecting the defect type and kicking off, and completing the automatic detection of the instant noodle packaging defects.
Further, the step a comprises:
(1) establishing an automatic detection device for the packaging defect of the instant noodles, collecting an image of the instant noodles on a conveyor belt by a camera, transmitting the image to a main control device, digitizing the image, and performing equalization processing on the gray value of the image by utilizing a histogram, wherein the gray value of the image after the equalization processing is as follows:
wherein, l is the gray level of the image, f (t) is the probability density function of the gray level of the image, t is a pixel point, the normalization processing is carried out on the gray level of the image, and when l belongs to [0,1], the probability density function value is 1; otherwise, the value is 0;
(2) if the gray value of the image pixel point (x, y) is g (x, y), the gray range is [ g [ ] 1 ，g 2 ]Mapping the gray value to g' (x, y) by a mapping function, and performing gray conversion:
where (α, β) is a gradation range after gradation conversion.
Further, the step B includes:
(1) color space processing of instant noodle packaging imagePerforming inter-conversion, performing initial segmentation on the image by a watershed method, and if a certain region is V r Neighborhood of V i (i ═ 1, 2.., n), then the similarity function is:
where ω is the weight coefficient, n is the number of neighborhoods, x i Is the mean value of the gray levels of each region,
obtaining a threshold value by a maximum inter-class variance method according to the gray level mean value of all the regions, wherein the similarity function value is larger than the threshold value, and the region is selected as a seed region;
(2) marking the selected seed region, traversing all seed regions and neighborhoods thereof, checking the neighborhood of the region if the seed region is not marked, and marking the same in the region if the neighborhood marks are the same; if the neighborhood marks are different, calculating the gray mean value difference between the region and the neighborhood, selecting the neighborhood with the minimum difference to carry out the same marking, traversing all the regions until all the regions are marked, combining adjacent pixels with similar gray values to the seed region into one region, combining a plurality of small fragmented regions in the segmented region, setting a region range threshold, and combining the regions smaller than the threshold into the neighborhood with the minimum gray mean value difference, thereby completing the image segmentation.
Further, the step C includes:
(1) constructing a Gaussian function of the instant noodle packaging image:
wherein, sigma is the standard deviation of the image normal distribution, (x, y) are the pixel coordinates, and (x) 0 ，y 0 ) Is the coordinates of the seed point, the scale space function of the image I (x, y) is:
S(x，y)＝h(x，y)*I(x，y)
wherein, the more the convolution is represented, the smaller the σ is, the smaller the image scale is, and the more obvious the detail feature is;
(2) the packaging image is convolved by a Gaussian function to generate a group of images, and the images are separated by a constant k in a scale space to obtain a Gaussian difference function:
D(x，y)＝[h(x，y)-h(k(x，y))]*I(x，y)
＝S(k(x，y))-S(x，y)
comparing pixel points in the image through a Gaussian difference function, setting a gray threshold, searching pixel points with gray values larger than the threshold, searching gray extreme points in the remaining pixel points as feature points to be selected, and performing Taylor expansion on the Gaussian difference function:
wherein, (x, y) T Is the offset of the pixel point, order
And finally, obtaining extreme points, removing pixel points at the edges in the candidate feature points, and extracting stable feature points.
Further, the step D includes:
calculating the Euclidean distance between the extracted characteristic image f (x, y) and the detection sample s (x, y):
wherein f is j And (x, y) is the jth characteristic, m is the number of the characteristics, min (d) and classification are carried out by searching the characteristic with the minimum Euclidean distance, the main control device counts the defect types and sends a signal to the kicking-off device for kicking off, and the package without the defect continues the next procedure, so that the automatic detection of the package defect of the instant noodles is completed.
The beneficial effects of the invention are:
under the condition that the packaging quality of the instant noodles is higher and higher, the method has better stability and robustness, reduces the omission factor and the false detection rate, improves the detection efficiency, has high detection speed, realizes nondestructive detection, saves labor resources, reduces the cost for enterprises, and makes the packaging of the instant noodles more refined.
Drawings
FIG. 1 is an overall flow chart of a machine vision inspection method for defects in instant noodle packaging;
FIG. 2 is a diagram of an automatic detection device for defects of instant noodles;
FIG. 3 is a flow chart of defect feature extraction for an image of an instant noodle package.
Detailed Description
Referring to fig. 1, the method of the present invention comprises the steps of:
A. establishing an automatic detection device model for the packaging defects of the instant noodles, collecting an image of the packaging of the instant noodles on a conveyor belt, preprocessing the image, and enhancing the image characteristics;
(1) and establishing an automatic detection device for the defects of the instant noodle package, as shown in figure 2. The camera gathers the instant noodle package image on the conveyer belt, conveys to master control set. The original image acquired in the field may have various noises, so that preprocessing is required. The image is digitized, and the histogram is used for carrying out equalization processing on the gray value of the image, so that the visual effect is enhanced. The gray scale of the image after the equalization processing is as follows:
wherein l is the gray level of the image, f (t) is the probability density function of the image gray level, and t is the pixel point. Normalizing the image gray level, and when l belongs to [0,1], the probability density function value is 1; otherwise it is 0.
(2) If the gray scale value of the image pixel point (x, y) is g (x, y), the gray scale range is [ g ] 1 ，g 2 ]Mapping the gray value to g' (x, y) by a mapping function, and performing gray conversion:
where (α, β) is a gradation range after gradation conversion. Thereby enlarging the gray scale range of the image and making the image clearer.
B. Selecting an image segmentation center pixel point, establishing a similarity measurement standard according to the continuity between pixel grayscales, and segmenting all pixels into different regions;
(1) and (4) performing color space conversion on the convenient surface packaging image, and performing initialization segmentation on the image by a watershed method. If a certain region is V r With a neighborhood of V i (i ═ 1, 2.., n), then the similarity function is:
where ω is the weight coefficient, n is the number of neighborhoods, x i Is the mean value of the gray levels of each region,
is the mean of the gray levels of all the regions. And obtaining a threshold value through a maximum inter-class variance method, wherein the similarity function value is larger than the threshold value, and the region is selected as a seed region.
(2) And marking the selected seed region, and traversing all the seed regions and the neighborhoods thereof. If the region is not marked, checking the neighborhood of the region, and if the neighborhood marks are the same, marking the region in the same way; if the adjacent domains are different in mark, the gray average value difference between the region and the adjacent domains is calculated, and the adjacent domains with the minimum difference are selected to carry out the same mark. All regions are traversed until all regions are marked. Thereby merging adjacent pixels having similar gray values as the seed region into one region. And setting a region range threshold value, and combining the regions smaller than the threshold value into a neighborhood with the minimum gray mean difference, thereby completing image segmentation.
C. Convolving the packaging image by a Gaussian kernel function to obtain a Gaussian difference image, and extracting the defect characteristics of the instant noodle packaging image by solving the gray extreme value of the image (as shown in figure 3);
(1) constructing a Gaussian function of the instant noodle packaging image:
wherein, sigma is the standard deviation of the image normal distribution, (x, y) is the pixel point coordinate, (x) 0 ，y 0 ) Are the coordinates of the seed points. The scale-space function of image I (x, y) is then:
S(x，y)＝h(x，y)*I(x，y)
where denotes convolution. The smaller sigma, the smaller the image scale, and the more prominent the detail features.
(2) The packaging image is convolved by a Gaussian function to generate a group of images, and the images are separated by a constant k in a scale space to obtain a Gaussian difference function:
D(x，y)＝[h(x，y)-h(k(x，y))]*I(x，y)
＝S(k(x，y))-S(x，y)
comparing pixel points in the image through a Gaussian difference function, setting a gray threshold, searching pixel points with gray values larger than the threshold, searching gray extreme points in the remaining pixel points as feature points to be selected, and performing Taylor expansion on the Gaussian difference function:
wherein, (x, y) T Is the offset of the pixel. Order to
Thereby obtaining an extreme point. And removing the pixel points at the edge in the candidate feature points, and extracting stable feature points.
D. Classifying by measuring the similarity between the defect characteristics and the sample, detecting the defect category and kicking off, and completing the automatic detection of the defects of the instant noodle packaging.
Calculating the Euclidean distance between the extracted characteristic image f (x, y) and the detection sample s (x, y):
wherein f is j And (x, y) is the jth characteristic, m is the number of the characteristics, min (d) and classification are carried out by searching the characteristic with the minimum Euclidean distance, the main control device counts the defect types and sends a signal to the kicking-off device for kicking off, and the package without the defect continues the next procedure, so that the automatic detection of the package defect of the instant noodles is completed.
In conclusion, the instant noodle packaging defect machine vision detection method is completed. The method has the advantages of good stability and robustness, low omission factor and false detection rate, high detection efficiency, high detection speed, nondestructive detection, labor resource saving, cost reduction for enterprises and more fine instant noodle packaging.
Claims (4)
1. A machine vision detection method for defects of instant noodle packages is characterized by comprising the following steps:
A. establishing an automatic detection device model for the packaging defects of the instant noodles, collecting an image of the packaging of the instant noodles on a conveyor belt, preprocessing the image, and enhancing the image characteristics;
(1) establishing an automatic detection device for the packaging defect of the instant noodles, collecting an image of the instant noodles on a conveyor belt by a camera, transmitting the image to a main control device, digitizing the image, and performing equalization processing on the gray value of the image by utilizing a histogram, wherein the gray value of the image after the equalization processing is as follows:
wherein, l is the gray level of the image, f (t) is the probability density function of the gray level of the image, t is a pixel point, the normalization processing is carried out on the gray level of the image, and when l belongs to [0,1], the probability density function value is 1; otherwise, the value is 0;
(2) if the gray value of the image pixel point (x, y) is g (x, y), the gray range is [ g [ ] 1 ,g 2 ]The gray scale value is mapped to g' (x, y) by a mapping function, and gray scale conversion is performed:
wherein (α, β) is a gray scale range after gray scale conversion;
B. selecting image segmentation center pixel points, establishing a similarity measurement standard according to the continuity between pixel grayscales, and segmenting all pixels into different regions;
C. convolving the segmented packaging image by a Gaussian kernel function to obtain a Gaussian difference image, and extracting the defect characteristics of the packaging image of the instant noodle by solving the gray extreme value of the image;
D. classifying by measuring the similarity between the defect characteristics and the sample, detecting the defect type and kicking off, and completing the automatic detection of the instant noodle packaging defects.
2. The instant noodle packaging defect machine vision inspection method of claim 1, wherein the step B comprises:
(1) color space conversion is carried out on the packing image of the instant noodles, the image is initially segmented by a watershed method, and if a certain area is V r Neighborhood of V i Where i is 1,2, …, n, the similarity function is:
where ω is the weight coefficient, n is the number of neighborhoods, x i Is the mean value of the gray levels of each region,
obtaining a threshold value by a maximum inter-class variance method according to the gray level mean value of all the regions, wherein the similarity function value is larger than the threshold value, and the region is selected as a seed region;
(2) marking the selected seed region, traversing all seed regions and neighborhoods thereof, checking the neighborhood of the region if the seed region is not marked, and marking the same in the region if the neighborhood marks are the same; if the neighborhood marks are different, calculating the gray mean value difference between the region and the neighborhood, selecting the neighborhood with the minimum difference to carry out the same marking, traversing all the regions until all the regions are marked, so that adjacent pixels with similar gray values to the seed region are combined into one region, a plurality of small fragmentary regions exist in the divided regions, setting a region range threshold value, and combining the regions smaller than the threshold value into the neighborhood with the minimum gray mean value difference, thereby completing image division.
3. The instant noodle packaging defect machine vision inspection method of claim 2, wherein the step C comprises:
(1) constructing a Gaussian function of the instant noodle package image:
wherein, sigma is the standard deviation of the image normal distribution, (x, y) is the pixel point coordinate, (x) 0 ,y 0 ) Is the coordinates of the seed point, the scale space function of the image I (x, y) is:
S(x,y)＝h(x,y)*I(x,y)
wherein, the more the convolution is represented, the smaller the σ is, the smaller the image scale is, and the more obvious the detail feature is;
(2) the packaging image is convolved by a Gaussian function to generate a group of images, and the images are separated by a constant k in a scale space to obtain a Gaussian difference function:
D(x,y)＝[h(x,y)-h(k(x,y))]*I(x,y)
＝S(k(x,y))-S(x,y)
comparing pixel points in the image through a Gaussian difference function, setting a gray threshold, searching pixel points with gray values larger than the threshold, searching gray extreme points in the rest pixel points as feature points to be selected, and carrying out Taylor expansion on the Gaussian difference function:
wherein, (x, y) T Is the offset of the pixel point, order
Therefore, an extreme point is obtained, pixel points located at the edge in the candidate feature points are removed, and stable feature points are extracted.
4. The machine vision inspection method of defects in instant noodle packaging according to claim 3 wherein step D comprises:
calculating the Euclidean distance between the extracted characteristic image f (x, y) and the detection sample s (x, y):
wherein f is j And (x, y) is the jth feature, m is the number of features, min (d) the feature with the minimum Euclidean distance is searched for classification, the main control device counts the defect types and sends a signal to the kicking-off device for kicking off, and the package without the defect continues the next procedure, so that the automatic detection of the package defect of the instant noodles is completed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811188773.1A CN109377485B (en) | 2018-10-12 | 2018-10-12 | Machine vision detection method for instant noodle packaging defects |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811188773.1A CN109377485B (en) | 2018-10-12 | 2018-10-12 | Machine vision detection method for instant noodle packaging defects |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109377485A CN109377485A (en) | 2019-02-22 |
| CN109377485B true CN109377485B (en) | 2022-09-06 |
Family
ID=65397514
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811188773.1A Active CN109377485B (en) | 2018-10-12 | 2018-10-12 | Machine vision detection method for instant noodle packaging defects |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109377485B (en) |
Families Citing this family (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112070710A (en) * | 2019-06-10 | 2020-12-11 | 北京平恒智能科技有限公司 | Method for detecting industrial product defect image |
| CN110415241A (en) * | 2019-08-02 | 2019-11-05 | 同济大学 | A computer vision-based surface quality detection method for concrete components |
| CN110717889A (en) * | 2019-09-06 | 2020-01-21 | 深圳新视智科技术有限公司 | Defect detection method and device based on digital printing, terminal and readable medium |
| CN111583084A (en) * | 2020-04-01 | 2020-08-25 | 杭州优视泰信息技术有限公司 | Workpiece defect positioning method |
| CN111968082A (en) * | 2020-07-30 | 2020-11-20 | 陕西科技大学 | Product packaging defect detection and identification method based on machine vision |
| KR102278895B1 (en) * | 2020-08-21 | 2021-07-20 | 주식회사농심 | detecting system for packing error |
| KR102259519B1 (en) * | 2020-08-28 | 2021-06-03 | 주식회사농심 | cutting error detecting system for pack unti |
| CN112862760B (en) * | 2021-01-19 | 2023-11-10 | 浙江大学 | Bearing outer ring surface defect area detection method |
| CN112862773A (en) * | 2021-01-29 | 2021-05-28 | 山东建筑大学 | Shaving board image defect classification detection method and device |
| CN113777033B (en) * | 2021-08-18 | 2024-08-02 | 长沙长泰机器人有限公司 | Raw strip defect detection method and device based on machine vision |
| CN113744239A (en) * | 2021-09-02 | 2021-12-03 | 广东奥普特科技股份有限公司 | Soft packet of lithium cell base angle automatic checkout device of polymer based on machine vision |
| CN113610851B (en) * | 2021-10-09 | 2021-12-31 | 武汉平桥兄弟包装材料有限公司 | Packaging decoration quality inspection method and system based on machine vision |
| CN114240888B (en) * | 2021-12-17 | 2022-09-20 | 沭阳安宏木制品厂 | Furniture assembly paint spraying defect repairing method and system based on image processing |
| CN114813794B (en) * | 2022-02-18 | 2023-10-03 | 成都飞机工业（集团）有限责任公司 | Method for acquiring scanning photo required by robot 3D printing nondestructive testing |
| CN114708325A (en) * | 2022-03-31 | 2022-07-05 | 南通华烨塑料工业有限公司 | Method for quickly positioning rubber production problem based on rubber blooming defect |
| CN114742788A (en) * | 2022-04-01 | 2022-07-12 | 南通高精数科机械有限公司 | Copper bar defect detection method and system based on machine vision |
| CN114706360A (en) * | 2022-06-06 | 2022-07-05 | 浙江新迪嘉禾食品有限公司 | Bread production quality control method and system |
| CN115311289A (en) * | 2022-10-12 | 2022-11-08 | 南通甜享纺织品有限公司 | Method for detecting oil stain defects of plain-color cloth |
| CN118428969B (en) * | 2023-09-18 | 2025-01-17 | 长春金融高等专科学校 | Method and system for tracing agricultural products by combining Internet of things |
| CN118365635B (en) * | 2024-06-17 | 2024-09-03 | 广东正一包装股份有限公司 | Visual inspection method and system for surface defects of packaging film |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330871A (en) * | 2017-06-29 | 2017-11-07 | 西安工程大学 | The image enchancing method of insulator automatic identification is run under bad weather condition |
| CN108416766A (en) * | 2018-01-31 | 2018-08-17 | 浙江理工大学 | Visual inspection method for defects of double-side incident light guide plate |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP5332181B2 (en) * | 2007-11-08 | 2013-11-06 | 味の素株式会社 | Packaging defect inspection equipment for packaging products with thin film |
| CN101470896A (en) * | 2007-12-24 | 2009-07-01 | 南京理工大学 | Automotive target flight mode prediction technique based on video analysis |
| CN202583085U (en) * | 2011-12-29 | 2012-12-05 | 天津普达软件技术有限公司 | Instant noodles sauce packet defect automatic detection device based on machine vision |
| CN102514767A (en) * | 2011-12-29 | 2012-06-27 | 天津普达软件技术有限公司 | Automatic detection device based on machine vision for defects of instant noodle sauce packets and method |
| CN103512888B (en) * | 2013-06-05 | 2015-11-11 | 北京化工大学 | A kind of cigarette packet seal defect detecting system based on image recognition technology |
| SG10201501672PA (en) * | 2015-03-05 | 2016-10-28 | Emage Vision Pte Ltd | Inspection of sealing quality in blister packages |
| CN104794491B (en) * | 2015-04-28 | 2018-01-23 | 重庆大学 | Based on the fuzzy clustering Surface Defects in Steel Plate detection method presorted |
| CN206914796U (en) * | 2017-07-07 | 2018-01-23 | 今麦郎面品有限公司 | A kind of improper packing product remover of instant noodles |
| CN107481241A (en) * | 2017-08-24 | 2017-12-15 | 太仓安顺财务服务有限公司 | A kind of color image segmentation method based on mixed method |
-
2018
- 2018-10-12 CN CN201811188773.1A patent/CN109377485B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107330871A (en) * | 2017-06-29 | 2017-11-07 | 西安工程大学 | The image enchancing method of insulator automatic identification is run under bad weather condition |
| CN108416766A (en) * | 2018-01-31 | 2018-08-17 | 浙江理工大学 | Visual inspection method for defects of double-side incident light guide plate |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109377485A (en) | 2019-02-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109377485B (en) | Machine vision detection method for instant noodle packaging defects | |
| CN115082419B (en) | Blow-molded luggage production defect detection method | |
| CN109961049B (en) | Cigarette brand identification method under complex scene | |
| CN109191459B (en) | Automatic identification and rating method for continuous casting billet macrostructure center segregation defect | |
| CN106056118B (en) | A kind of identification method of counting for cell | |
| CN104990925B (en) | One kind is based on gradient multi thresholds optimization defect inspection method | |
| CN111310760B (en) | Oracle Bone Inscription Text Detection Method Combining Local Prior Features and Deep Convolution Features | |
| CN103048329B (en) | A kind of road surface crack detection method based on active contour model | |
| CN114972356B (en) | Plastic product surface defect detection and identification method and system | |
| CN107610114A (en) | Optical satellite remote sensing image cloud snow mist detection method based on SVMs | |
| CN110070523B (en) | A kind of foreign matter detection method for bottle bottom | |
| CN110648330B (en) | Defect detection method for camera glass | |
| CN110175556B (en) | Remote sensing image cloud detection method based on Sobel operator | |
| CN114820625B (en) | Automobile top block defect detection method | |
| CN113221881B (en) | A multi-level smartphone screen defect detection method | |
| CN106097368A (en) | A kind of recognition methods in veneer crack | |
| CN112734761A (en) | Industrial product image boundary contour extraction method | |
| CN107516315B (en) | A method for monitoring slag discharge of roadheader based on machine vision | |
| CN114511770A (en) | Road sign plate identification method | |
| CN111968082A (en) | Product packaging defect detection and identification method based on machine vision | |
| CN110060239B (en) | A kind of defect detection method for bottle mouth | |
| CN112734729A (en) | Water gauge water level line image detection method and device suitable for night light supplement condition and storage medium | |
| Fang et al. | Towards real-time crack detection using a deep neural network with a Bayesian fusion algorithm | |
| CN114550167B (en) | Artificial intelligence based pear quality classification method and device | |
| CN103116890A (en) | Video image based intelligent searching and matching method |
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 | ||
| TA01 | Transfer of patent application right | ||
| TA01 | Transfer of patent application right |
Effective date of registration: 20220817 Address after: 518000 Room 2107, No. 51, Pingxin North Road, Shangmu Ancient Community, Pinghu Street, Longgang District, Shenzhen, Guangdong, China Applicant after: Shenzhen Zhongzhi Vision Technology Co., Ltd. Address before: 265700 No. 218 Huancheng North Road, Donglai Street, Longkou City, Yantai City, Shandong Province Applicant before: LONGKOU WEIMEISI ENVIRONMENTAL PROTECTION TECHNOLOGY Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |