CN112215852A - Digital image segmentation method based on cluster learning device integration - Google Patents
Digital image segmentation method based on cluster learning device integration Download PDFInfo
- Publication number
- CN112215852A CN112215852A CN202011049482.1A CN202011049482A CN112215852A CN 112215852 A CN112215852 A CN 112215852A CN 202011049482 A CN202011049482 A CN 202011049482A CN 112215852 A CN112215852 A CN 112215852A
- Authority
- CN
- China
- Prior art keywords
- digital image
- digital
- learner
- rgb
- screening
- 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 24
- 230000010354 integration Effects 0.000 title claims abstract description 17
- 238000003709 image segmentation Methods 0.000 title claims abstract description 12
- 238000012216 screening Methods 0.000 claims abstract description 21
- 230000011218 segmentation Effects 0.000 claims abstract description 10
- 238000004364 calculation method Methods 0.000 claims abstract description 8
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000009499 grossing Methods 0.000 claims description 4
- 239000003086 colorant Substances 0.000 claims 1
- 230000000694 effects Effects 0.000 abstract description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
Classifications
-
- 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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
-
- 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
- G06T2207/20081—Training; Learning
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a digital image segmentation method based on clustering learner integration, which relates to the field of digital image processing and specifically comprises the following steps: the method comprises the following steps: inputting digital image information into a digital image processing device; step two: carrying out rough segmentation on the digital image; step three: finely dividing the digital image; step four: generating a clustering learner ensemble; step five: setting RGB threshold values; step six: the digital image identified and screened in the fifth step is searched and matched with the image information in the database; step seven: and outputting the digital image obtained after matching. In the implementation process of the invention, the clustering learner is utilized to improve the segmentation precision of the digital image, and when the image is segmented, useless parts are removed preferentially, so that the effects of reducing the calculation load of hardware and improving the segmentation efficiency are achieved; meanwhile, the coordinate system is used as the specific representation of screening, so that the screening can be intuitively understood during screening, and the screening efficiency is improved.
Description
Technical Field
The invention relates to the field of digital image processing, in particular to a digital image segmentation method based on clustering learner integration.
Background
With the popularization of digital image devices such as digital cameras, digital images have been widely used in various industries. In order to effectively extract and utilize information contained in a digital image, the image needs to be segmented. The image segmentation means that regions with different semantics in an image are segmented, the regions are not intersected with each other, and each region meets the consistency. The effective digital image segmentation technology lays a foundation for further digital image retrieval, identification and the like. Cluster learners are classic machine learning techniques.
When the cluster learner is integrated, the whole digital image needs to be processed, and in the process, a large amount of useless calculation needs to be carried out, so that the operation burden of machine hardware is greatly increased, and the efficiency is low.
Disclosure of Invention
The invention aims to provide a digital image segmentation method based on clustering learner integration, so as to solve the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme:
the digital image segmentation method based on the cluster learner integration comprises the following steps:
the method comprises the following steps: inputting the digital image information into a digital image processing device, and preprocessing the digital image information by the digital image processing device; the preprocessing comprises smoothing and denoising processing, so that the overall quality of the digital picture is improved;
step two: roughly dividing the digital image, and leaving the image which needs to be used and removing the image which does not need to be used;
step three: finely dividing the digital image;
step four: generating a clustering learner integration, and carrying out primary classification and screening on the information of each digital image through RGB identification to obtain a segmentation result;
step five: setting RGB threshold values, further screening the digital images screened in the step four, and screening the digital images in the RGB threshold values, so that the retrieval precision is improved;
step six: the digital image identified and screened in the fifth step is searched and matched with the image information in the database;
step seven: and outputting the digital image obtained after matching.
As a further scheme of the invention: when the digital image is roughly divided in the second step, the target area to be processed is cut out from the whole digital image through manual operation through a computer, so that interference of unnecessary images to the fourth step is avoided.
As a further scheme of the invention: in the third step, when the digital image is subdivided, it is necessary to ensure that the areas of the divided regions of each digital image are the same.
As a further scheme of the invention: and integrating the clustering learner in the fourth step into each segmented digital image when generating, wherein all the images containing the target color are selected, and the rest digital images are eliminated.
As a further scheme of the invention: and when the RGB threshold value is set in the step five, obtaining a screening result in the following mode:
(1) establishing an RGB threshold coordinate system, taking R (red), G (green) and B (blue) as coordinate axes in the three-dimensional coordinate system respectively, and forming a square range in the coordinate system by using RGB thresholds respectively;
(2) calculating the ratio of RGB in each digital image, setting the area occupied by RGB in the digital image as Sn during calculation, setting the total area of the digital image as S, obtaining the ratio of Sn to S, and respectively marking the ratio in an RGB threshold coordinate system;
(3) and rejecting unqualified digital images, and paying attention to at least one point of the digital image in the square body during processing.
Compared with the prior art, the invention has the beneficial effects that:
in the implementation process of the invention, the clustering learner is utilized to improve the segmentation precision of the digital image, and when the image is segmented, useless parts are removed preferentially, so that the effects of reducing the calculation load of hardware and improving the segmentation efficiency are achieved; meanwhile, the coordinate system is used as the specific representation of screening, so that the screening can be intuitively understood during screening, and the screening efficiency is improved.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
In the embodiment of the invention, a digital image segmentation method based on cluster learner integration comprises the following steps:
the method comprises the following steps: inputting the digital image information into a digital image processing device, and preprocessing the digital image information by the digital image processing device; the pre-processing includes smoothing and denoising processes to improve the overall quality of the digital picture.
Step two: the digital image is roughly divided, the images which need to be used are left, and the images which do not need to be used are removed, so that the calculation amount during the digital image processing can be effectively reduced, and the processing efficiency is improved.
Step three: the digital image is finely divided, and when the digital image is finely divided, the areas of the divided regions of each digital image need to be ensured to be the same.
Step four: and generating a clustering learner integration, and carrying out primary classification and screening on the information of each digital image through RGB identification to obtain a segmentation result.
Step five: and setting RGB threshold values, further screening the digital images screened in the step four, and screening the digital images in the RGB threshold values, thereby improving the retrieval precision.
Step six: and fifthly, identifying the screened digital images in the step five, retrieving the images, and matching the images with image information in a database.
Step seven: and outputting the digital image obtained after matching.
Preferably, when the digital image is roughly divided in the second step, the target area to be processed is cut out from the whole digital image by manual operation through a computer, so that interference of unnecessary images on the fourth step is avoided.
Preferably, the cluster learner in step four is integrated in the generation process, all the images containing the target color in each segmented digital image are selected, and all the rest digital images are excluded.
Preferably, when the RGB threshold is set in the fifth step, the screening result is obtained as follows:
(1) establishing an RGB threshold coordinate system, taking R (red), G (green) and B (blue) as coordinate axes in the three-dimensional coordinate system respectively, and forming a square range in the coordinate system by using RGB thresholds respectively;
(2) calculating the ratio of RGB in each digital image, setting the area occupied by RGB in the digital image as Sn during calculation, setting the total area of the digital image as S, obtaining the ratio of Sn to S, and respectively marking the ratio in an RGB threshold coordinate system;
(3) and rejecting unqualified digital images, and paying attention to at least one point of the digital image in the square body during processing.
Example 2
In the embodiment of the invention, the license plate recognition is taken as an example, and the method comprises the following steps:
the method comprises the following steps: inputting the digital image shot by the camera into a digital image processing device, and preprocessing the digital image by the digital image processing device; the preprocessing comprises smoothing and denoising processing, and the integral definition of the license plate is improved.
Step two: and carrying out rough segmentation on the digital image, manually intercepting a part of a target license plate in the digital image to be used as a target processing image, and reducing the calculation amount during hardware image processing.
Step three: the digital image is finely divided, and when the digital image is finely divided, the areas of the divided regions of each digital image need to be ensured to be the same.
Step four: and generating a clustering learning device integration, screening out all the white parts containing the blue part and the font of the license plate in the digital image, and excluding all the parts except the license plate.
Step five: and setting RGB threshold values, mainly aiming at the blue part of the license plate, and removing the blue part when the occupation ratio of the blue part occupying the digital image exceeds the set threshold value, wherein the rest part is the font part of the license plate.
Step six: and fifthly, identifying the screened digital images in the step five, retrieving the images, and matching the images with image information in a database.
Step seven: and outputting the digital image obtained after matching.
Although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that various changes in the embodiments and/or modifications of the invention can be made, and equivalents and modifications of some features of the invention without departing from the spirit and scope of the invention.
Claims (5)
1. A digital image segmentation method based on cluster learning device integration is characterized by comprising the following steps:
the method comprises the following steps: inputting the digital image information into a digital image processing device, and preprocessing the digital image information by the digital image processing device; the preprocessing comprises smoothing and denoising processing, so that the overall quality of the digital picture is improved;
step two: roughly dividing the digital image, and leaving the image which needs to be used and removing the image which does not need to be used;
step three: finely dividing the digital image;
step four: generating a clustering learner integration, and carrying out primary classification and screening on the information of each digital image through RGB identification to obtain a segmentation result;
step five: setting RGB threshold values, further screening the digital images screened in the step four, and screening the digital images in the RGB threshold values, so that the retrieval precision is improved;
step six: the digital image identified and screened in the fifth step is searched and matched with the image information in the database;
step seven: and outputting the digital image obtained after matching.
2. The digital image segmentation method based on cluster learner integration according to claim 1, wherein when the digital image is roughly segmented in the second step, the target area to be processed is cut out from the whole digital image through manual operation of a computer, so that interference of unnecessary images to the fourth step is avoided.
3. The method for segmenting digital images based on the integration of a cluster learner as claimed in claim 1, wherein in the third step, when performing the segmentation of the digital images, it is required to ensure that the segmented areas of each digital image are the same in size.
4. The method for segmenting digital images based on cluster learner integration according to claim 1, wherein the cluster learner integration in step four is used for selecting all images containing target colors and excluding all other digital images in each segmented digital image during generation.
5. The digital image segmentation method based on cluster learner integration according to claim 1, wherein when the RGB threshold is set in the fifth step, the screening result is obtained by:
(1) establishing an RGB threshold coordinate system, taking R (red), G (green) and B (blue) as coordinate axes in the three-dimensional coordinate system respectively, and forming a square range in the coordinate system by using RGB thresholds respectively;
(2) calculating the ratio of RGB in each digital image, setting the area occupied by RGB in the digital image as Sn during calculation, setting the total area of the digital image as S, obtaining the ratio of Sn to S, and respectively marking the ratio in an RGB threshold coordinate system;
(3) and rejecting unqualified digital images, and paying attention to at least one point of the digital image in the square body during processing.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011049482.1A CN112215852A (en) | 2020-09-29 | 2020-09-29 | Digital image segmentation method based on cluster learning device integration |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011049482.1A CN112215852A (en) | 2020-09-29 | 2020-09-29 | Digital image segmentation method based on cluster learning device integration |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112215852A true CN112215852A (en) | 2021-01-12 |
Family
ID=74051386
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011049482.1A Pending CN112215852A (en) | 2020-09-29 | 2020-09-29 | Digital image segmentation method based on cluster learning device integration |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112215852A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113139975A (en) * | 2021-04-19 | 2021-07-20 | 国交空间信息技术（北京）有限公司 | Road characteristic-based road surface segmentation method and device |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7630544B1 (en) * | 2005-04-06 | 2009-12-08 | Seiko Epson Corporation | System and method for locating a character set in a digital image |
| CN102496019A (en) * | 2011-12-08 | 2012-06-13 | 银江股份有限公司 | License plate character segmenting method |
| US20170300786A1 (en) * | 2015-10-01 | 2017-10-19 | Intelli-Vision | Methods and systems for accurately recognizing vehicle license plates |
| CN108596920A (en) * | 2018-05-02 | 2018-09-28 | 北京环境特性研究所 | A kind of Target Segmentation method and device based on coloured image |
| CN109447998A (en) * | 2018-09-29 | 2019-03-08 | 华中科技大学 | Based on the automatic division method under PCANet deep learning model |
| CN111339953A (en) * | 2020-02-27 | 2020-06-26 | 广西大学 | Clustering analysis-based mikania micrantha monitoring method |
-
2020
- 2020-09-29 CN CN202011049482.1A patent/CN112215852A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7630544B1 (en) * | 2005-04-06 | 2009-12-08 | Seiko Epson Corporation | System and method for locating a character set in a digital image |
| CN102496019A (en) * | 2011-12-08 | 2012-06-13 | 银江股份有限公司 | License plate character segmenting method |
| US20170300786A1 (en) * | 2015-10-01 | 2017-10-19 | Intelli-Vision | Methods and systems for accurately recognizing vehicle license plates |
| CN108596920A (en) * | 2018-05-02 | 2018-09-28 | 北京环境特性研究所 | A kind of Target Segmentation method and device based on coloured image |
| CN109447998A (en) * | 2018-09-29 | 2019-03-08 | 华中科技大学 | Based on the automatic division method under PCANet deep learning model |
| CN111339953A (en) * | 2020-02-27 | 2020-06-26 | 广西大学 | Clustering analysis-based mikania micrantha monitoring method |
Non-Patent Citations (1)
| Title |
|---|
| 明月星;马静;张杨;: "基于K-means算法车牌分割的应用", 科技视界, no. 03 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113139975A (en) * | 2021-04-19 | 2021-07-20 | 国交空间信息技术（北京）有限公司 | Road characteristic-based road surface segmentation method and device |
| CN113139975B (en) * | 2021-04-19 | 2023-11-17 | 国交空间信息技术（北京）有限公司 | Road feature-based pavement segmentation method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104751142B (en) | A kind of natural scene Method for text detection based on stroke feature | |
| DE102020100684B4 (en) | MARKING OF GRAPHICAL REFERENCE MARKERS | |
| CN109145915B (en) | Rapid distortion correction method for license plate under complex scene | |
| CN109886896B (en) | Blue license plate segmentation and correction method | |
| CN108491498B (en) | Bayonet image target searching method based on multi-feature detection | |
| CN105678213B (en) | Dual-mode mask person event automatic detection method based on video feature statistics | |
| CN106803257B (en) | Method for segmenting disease spots in crop disease leaf image | |
| CN103971128A (en) | Traffic sign recognition method for driverless car | |
| CN109961016B (en) | Multi-gesture accurate segmentation method for smart home scene | |
| CN106778779A (en) | A kind of electric injection molding machine mould detection method | |
| CN111695373B (en) | Zebra stripes positioning method, system, medium and equipment | |
| CN109145964B (en) | Method and system for realizing image color clustering | |
| CN110826408A (en) | Face recognition method by regional feature extraction | |
| CN107146258B (en) | Image salient region detection method | |
| CN105740844A (en) | Insulator cracking fault detection method based on image identification technology | |
| CN112215852A (en) | Digital image segmentation method based on cluster learning device integration | |
| CN111429376A (en) | High-efficiency digital image processing method with high-precision and low-precision integration | |
| CN107145888A (en) | Video caption real time translating method | |
| CN113840135A (en) | Color cast detection method, device, equipment and storage medium | |
| CN114066993A (en) | Power distribution cabinet control panel segmentation method based on machine vision | |
| CN109934152B (en) | Improved small-bent-arm image segmentation method for sign language image | |
| CN109766860B (en) | Face detection method based on improved Adaboost algorithm | |
| WO2019041447A1 (en) | 3d video frame feature point extraction method and system | |
| CN117765287A (en) | Image target extraction method combining LWR and density clustering | |
| CN114926635B (en) | Target segmentation method in multi-focus image combined with deep learning 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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20210112 |