CN109241390A - A kind of intelligent design system and method for the decorative textile product based on big data - Google Patents
A kind of intelligent design system and method for the decorative textile product based on big data Download PDFInfo
- Publication number
- CN109241390A CN109241390A CN201811014178.6A CN201811014178A CN109241390A CN 109241390 A CN109241390 A CN 109241390A CN 201811014178 A CN201811014178 A CN 201811014178A CN 109241390 A CN109241390 A CN 109241390A
- Authority
- CN
- China
- Prior art keywords
- image
- unit
- sample image
- color
- textile product
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2113/00—Details relating to the application field
- G06F2113/12—Cloth
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
A kind of intelligent design system and method for the decorative textile product based on big data, it includes image collection unit, storage unit, image processing unit, characteristic statistics unit and Combination Design unit, large database concept is formed by the sample image that different channels collect decorative textile product, the pattern and color characteristic of different sample images in large database concept are extracted using image processing unit, and pass through the pattern and the color characteristic frequency of the statistics extraction of characteristic statistics unit, the part colours feature for sorting forward is automatically selected by Combination Design unit and partial pattern feature auto composition generates decorative textile product composition typesetting scheme, realize the elemental characteristic that fashion trend instantly is obtained using big data, and realize the intelligent design of decorative textile product, substantially increase design efficiency.
Description
Technical field
The present invention relates to a kind of designing system of textile and methods, and in particular, to a kind of decoration based on big data
With the intelligent design system and method for textile.
Background technique
Decorative textile product belongs to the scope of utility art, has both practicability and decorative dual function.Decoration spinning
The utility function of fabric mainly passes through the inherent quality of product and technical performance is realized, decoration functions are then mainly to pass through
Pattern, color, style, the material of product etc. are reflected.The two organically combines, and embodies comprehensive group effect.
The design of decorative textile product is created different from general plastic arts, it is not only the high art effect appreciated for people, and
Practical textile for beautifying the environment.The pattern of decorative textile product largely decides the value of the textile,
The soul of textile can be called.The design level of pattern not only influences the appearance and its artistic effect of silk fabrics, or even also
It will affect the quality and economic benefit of fabric.Therefore, the design needs of pattern continue to introduce new, and instantly popular set is added
Element or theory are counted, to adapt to current continually changing demand.
In the prior art, the fashion trend prediction of textile, mainly by consumer, representative user, designer etc.
Investigation and test marketing.Either in the formation of textile and explaination stage still in information collection, the analysis of fashion trend
And the refinement stage largely relies on the analytical judgment of people, lacks data and supports.Design to textile is also mainly people
Increase certain fashion trend elements selectively on the basis of existing textile for ground to complete design, efficiency is lower, is not easy to
It quickly updates, and the pattern of textile and color have been difficult to biggish breakthrough.
Summary of the invention
The present invention provides a kind of decorative textile product based on big data to solve the above problem in the prior art
Intelligent design system and method, to utilize raising design efficiency.
Specifically, the invention is realized by the following technical scheme:
The present invention is based on the intelligent design systems of the decorative textile product of big data, comprising: image collection unit, storage unit,
Image processing unit, characteristic statistics unit and Combination Design unit.
Described image collector unit is used to collect the sample image of different decorative textile products, and the sample image can lead to
Cross the source such as network search engines material database, commercial distribution website, retail market, related meeting for the placement of orders or major information platform
It collects.
Described image processing unit is used to extract multiple features of sample image described in storage unit, the multiple feature
Design and color including textile.
Described image processing unit further comprises image pre-processing unit and feature extraction unit.
The sample image is smoothed by described image pretreatment unit, the smoothing processing is in Gauss
According to the sum weight of space length, color distance and neighborhood gradient distance adjustment neighborhood territory pixel point to construct on the basis of filtering
Filter factor, then linear filtering is carried out to each pixel of the sample image.By image preprocessing, to inhibit to weave
The influence that the information unrelated with pattern such as product texture extracts pattern.
The feature extraction unit carries out gray processing processing to the sample image after the smoothing processing using mean value method,
And image is split using edge detection operator using the method for automatically selecting discrimination threshold, to obtain the sample graph
The pattern characteristics of picture.Edge detection operator can detecte the weak edge that many brief summaries in textile images are configured to, thus
If these edges can be divided out substantially when pattern extracts, and clear-cut, accuracy is high, therefore is mentioned using which
The decorative textile product pattern edge effect taken is best, and has the advantages such as translation scaling and rotation invariant.
The feature extraction unit also extracts rgb value, common face by the sample image stored from the storage unit
The colour space has: RGB, HSV, HSI, HSL and CMYK etc., and using which kind of color space, there is no unified marks in the prior art
It is quasi-.Wherein, common RGB color model is to indicate various colors with three kinds of primary colours of red, green, blue, but RGB color is not
It can combine well with color space perceptually, preferably HSV model, therefore need RGB to HSV's herein
Conversion.The rgb value is converted to the hsv color space for being suitble to naked eyes to differentiate by the present invention, and counts the picture of each color component
Prime number accounts for the ratio of image total pixel number, and to obtain the ratio distribution of image various colors component, the ratio distribution is used for table
Up to color characteristic.Ratio distribution is the function of colouring information, it indicates for having with the other pixel of color level in image
Number, abscissa are color ranks, and ordinate is the frequency that color occurs.Using the color feature extracted mode, even if right
In complex background image, also there is relatively good robustness, and do not influenced by picture size and direction.
The characteristic statistics unit counts homogenous characteristics based on the textile color characteristic and pattern characteristics of extraction respectively and goes out
The existing frequency, and the color characteristic and pattern characteristics are ranked up based on the frequency.
One or more color characteristics for sorting forward in characteristic statistics unit described in the Combination Design Unit selection and
Pattern characteristics carry out composition of geometry, and generate decorative textile product composition typesetting scheme preview original text.Specifically, sort forward number
Amount can be reasonably selected according to calculating with storage capacity, and preceding 100,200 or 500 feature that for example, sorts can be used as combination and set
Count the source of composition.
Further, according to probabilistic model, the matching probability between different colours feature and different pattern characteristics, choosing are calculated
Most suitable one or more colors-pattern characteristics combination is selected, ornament textile designs scheme is formed.And user can be with root
The combination of each feature is carried out from main modulation according to individual demand.
The Intelligentized design method of decorative textile product based on big data of the invention, comprising the following steps:
The sample image of different decorative textile products is collected, the sample image can be for example, by network search engines material
Library, commercial distribution website, retail market, the related sources such as meeting for the placement of orders or major information platform are collected.
The sample image is stored, and gradually accumulation forms large database concept.
Multiple features of the sample image of storage are extracted, the multiple feature includes the face of the decorative textile product
Color characteristic and pattern characteristics；Specifically, the sample image is smoothed, the smoothing processing is in gaussian filtering
On the basis of according to the sum weight of space length, color distance and neighborhood gradient distance adjustment neighborhood territory pixel point to construct filtering system
Number, then linear filtering is carried out to each pixel of the sample image.By image preprocessing, to inhibit textile texture
The influence that pattern is extracted Deng the information unrelated with pattern.
Further, the present invention carries out gray processing processing to the sample image after the smoothing processing using mean value method,
And image is split using edge detection operator using the method for automatically selecting discrimination threshold, to obtain the sample graph
The pattern characteristics of picture；Edge detection operator can detecte the weak edge that many brief summaries in textile images are configured to, thus
If these edges can be divided out substantially when pattern extracts, and clear-cut, accuracy is high, therefore is mentioned using which
The decorative textile product pattern edge effect taken is best, and has the advantages such as translation scaling and rotation invariant.
Rgb value is also extracted by the sample image of storage, the rgb value is converted to the hsv color for being suitble to naked eyes to differentiate
Space, and the pixel number for counting each color component accounts for the ratio of image total pixel number, to obtain image various colors component
Ratio distribution, the ratio distribution is for expressing color characteristic.Ratio distribution is the function of colouring information, it indicates image
In there is number with the other pixel of color level, abscissa is color rank, and ordinate is the frequency that color occurs.Using
The color feature extracted mode, even for complex background image, it may have relatively good robustness, and not by image
The influence of size and direction.
Textile color characteristic and pattern characteristics based on extraction count the frequency of homogenous characteristics appearance respectively, and are based on institute
The frequency is stated to be ranked up the color characteristic and pattern characteristics.
The part colours feature for sorting forward in the characteristic statistics unit and partial pattern feature is selected to carry out geometry structure
Figure, and generate decorative textile product composition typesetting scheme preview original text.Specifically, the forward quantity that sorts can be according to calculating and deposit
Energy storage power reasonably selects, and preceding 100,200 or 500 feature that for example, sorts can be used as the source of Combination Design composition.
Alternatively, based on the considerations of the aspect and computing capability of data volume storage, sequence can also be leaned on
Preceding pattern characteristics and color characteristic selectively deposit image feature base, thus only need to be from characteristics of image in designing picture composition
Suitable color-pattern characteristics combination is matched in database to generate decorative textile product design scheme.
By using technical solution of the present invention, the beneficial effect is that:
(1) big data era brings the change that information such as collects and analyzes at the working methods, and the present invention is exactly logical by computer
It crosses based on multiple channel collects the big data that the sample images of different decorative textile products is constituted and carries out characteristic statistics point
Analysis is well reflected instantly popular hot spot and trend, is accurately positioned popular elemental characteristic, realizes data and see clearly, thus
Key foundation is provided for science, intelligentized design.
(2) present invention effectively inhibits textile texture etc. unrelated with pattern by pre-processing to sample image
The influence that information extracts pattern；The pattern characteristics profile that specific image zooming-out and color extraction mode obtain through the invention
Clearly, accuracy is high, and effect is good, and has the advantages such as translation scaling and rotation invariant, and color characteristic is not also by image
The influence of size and direction.
(3) present invention realizes Auto-matching color-pattern characteristics to obtain decorative textile product using by different modes
Design scheme not only enriches design pattern, and substantially increases designer and designing textile patter and when color
Efficiency, convenient for quickly updating design scheme.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to embodiments to the present invention
It is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not used to
Limit the present invention.
Decorative textile product of the invention includes but is not limited to following several classes: ground is laid with class such as carpet, ground cushion etc.；Hang curtain
Hide decorations such as curtain, door curtain, screen；Furniture, household electrical appliances cover decorations class such as tablecloth, slip cover etc.；Metope patch decorations class such as wall paper；It is used on bed
Category, sanitary toilet class such as towel bath towel etc., the miscellaneous decorations class of kitchen etc..
The present invention provides a kind of intelligent design system of decorative textile product based on big data, comprising: image collection list
Member, storage unit, image processing unit, characteristic statistics unit and Combination Design unit.
Described image collector unit is used to collect the sample image of different decorative textile products, and the sample image can lead to
Cross the source such as network search engines material database, commercial distribution website, retail market, related meeting for the placement of orders or major information platform
It collects.
The storage unit is used to store the sample image of described image collector unit, and gradually accumulation forms big data
Library.
Described image processing unit is used to extract multiple features of sample image described in storage unit, the multiple feature
Design and color including textile.
Described image processing unit further comprises image pre-processing unit and feature extraction unit.
The sample image is smoothed by described image pretreatment unit, the smoothing processing is in Gauss
According to the sum weight of space length, color distance and neighborhood gradient distance adjustment neighborhood territory pixel point to construct on the basis of filtering
Filter factor, then linear filtering is carried out to each pixel of the sample image.By image preprocessing, to inhibit to weave
The influence that the information unrelated with pattern such as product texture extracts pattern.
The feature extraction unit carries out gray processing processing to the sample image after the smoothing processing using mean value method,
And image is split using edge detection operator using the method for automatically selecting discrimination threshold, to obtain the sample graph
The pattern characteristics of picture.Edge detection operator can detecte the weak edge that many brief summaries in textile images are configured to, thus
If these edges can be divided out substantially when pattern extracts, and clear-cut, accuracy is high, therefore is mentioned using which
The decorative textile product pattern edge effect taken is best, and has the advantages such as translation scaling and rotation invariant.
The feature extraction unit also extracts rgb value by the sample image stored from the storage unit, by the RGB
Value is converted to the hsv color space for being suitble to naked eyes to differentiate, and the pixel number for counting each color component accounts for image total pixel number
Ratio, to obtain the ratio distribution of image various colors component, the ratio distribution is for expressing color characteristic.The ratio point
Cloth is the function of colouring information, it indicates the number for having with the other pixel of color level in image, and abscissa is color level
Not, ordinate is the frequency that color occurs.Also had using the color feature extracted mode even for complex background image
There is relatively good robustness, and is not influenced by picture size and direction.
The characteristic statistics unit counts homogenous characteristics based on the textile color characteristic and pattern characteristics of extraction respectively and goes out
The existing frequency, and the color characteristic and pattern characteristics are ranked up based on the frequency.
One or more color characteristics for sorting forward in characteristic statistics unit described in the Combination Design Unit selection and
Pattern characteristics carry out composition of geometry, and generate decorative textile product composition typesetting scheme preview original text.Specifically, sort forward number
Amount can be reasonably selected according to calculating with storage capacity, and preceding 100,200 or 500 feature that for example, sorts can be used as combination and set
Count the source of composition.
Further, according to probabilistic model, the matching probability between different colours feature and different pattern characteristics, choosing are calculated
Most suitable one or more colors-pattern characteristics combination is selected, ornament textile designs scheme is formed.And user can be with root
The combination of each feature is carried out from main modulation according to individual demand.
Specifically, the present invention is based on the Intelligentized design methods of the decorative textile product of big data, comprising the following steps:
S1, the sample image for collecting different decorative textile products, the sample image can be for example, by network search engines elements
Material library, commercial distribution website, retail market, the related sources such as meeting for the placement of orders or major information platform are collected.
S2, the storage sample image, and gradually accumulation forms large database concept.
S3, the multiple features for extracting the sample image stored, the multiple feature includes the decorative textile product
Color characteristic and pattern characteristics；Further comprise: the sample image is smoothed, the smoothing processing is in height
According to the sum weight of space length, color distance and neighborhood gradient distance adjustment neighborhood territory pixel point with structure on the basis of this filtering
Filter factor is built, then linear filtering is carried out to each pixel of the sample image.Using mean value method to the smooth place
Sample image after reason carries out gray processing processing, and utilizes edge detection operator to figure using the method for automatically selecting discrimination threshold
As being split, to obtain the pattern characteristics of the sample image.And rgb value is also extracted by the sample image of storage,
The rgb value is converted to the HSV color space for being suitble to naked eyes to differentiate, and the pixel number for counting each color component accounts for image
The ratio of total pixel number, to obtain the ratio distribution of image various colors component.
S4, the textile color characteristic based on extraction and pattern characteristics count the frequency of homogenous characteristics appearance, and base respectively
The color characteristic and pattern characteristics are ranked up in the frequency.
The part colours feature and partial pattern feature for sorting forward in S5, the selection characteristic statistics unit carry out geometry
Composition, and generate decorative textile product composition typesetting scheme preview original text.
Claims (4)
1. a kind of intelligent design system of the decorative textile product based on big data, which is characterized in that
It include: image collection unit, storage unit, image processing unit, characteristic statistics unit and Combination Design unit；
Described image collector unit is used to collect the sample image of different decorative textile products；
The storage unit is used to store the sample image of described image collector unit, and gradually accumulation forms large database concept；
Described image processing unit is used to extract multiple features of sample image described in storage unit, and the multiple feature includes
The design and color of textile；
Described image processing unit further comprises image pre-processing unit, feature extraction unit；It is pre-processed by described image
Unit is smoothed the sample image, then carries out linear filtering to each pixel of the sample image；
The feature extraction unit carries out gray processing processing to the sample image after the smoothing processing using mean value method, and adopts
Image is split using edge detection operator with the method for automatically selecting discrimination threshold, to obtain the sample image
Pattern characteristics；
The feature extraction unit also extracts rgb value by the sample image stored from the storage unit, and the rgb value is turned
It is changed to the HSV color space for being suitble to naked eyes to differentiate, and the pixel number for counting each color component accounts for the ratio of image total pixel number
Example, to obtain the ratio distribution of image various colors component, the ratio distribution is for expressing color characteristic；
The characteristic statistics unit counts homogenous characteristics appearance based on the textile color characteristic and pattern characteristics of extraction respectively
The frequency, and the color characteristic and pattern characteristics are ranked up based on the frequency；
Sort forward one or more color characteristics and pattern in characteristic statistics unit described in the Combination Design Unit selection
Feature carries out composition of geometry, and generates decorative textile product composition typesetting scheme preview original text.
2. intelligent design system according to claim 1, which is characterized in that the smoothing processing is the base in gaussian filtering
According to the sum weight of space length, color distance and neighborhood gradient distance adjustment neighborhood territory pixel point to construct filtering system on plinth
Number.
3. intelligent design system according to claim 1, which is characterized in that the sample image passes through network search engines
Material database, commercial distribution website, retail market, the related sources such as meeting for the placement of orders or major information platform are collected.
4. a kind of decorative textile product intelligent design system based on described in any one of claims 1 to 3 based on big data
Method, which comprises the following steps:
The sample image of different decorative textile products is collected, the sample image is sold by network search engines material database, business
Sell website, retail market, the related sources such as meeting for the placement of orders or major information platform collection；
The sample image is stored, and gradually accumulation forms large database concept；
Multiple features of the sample image of storage are extracted, the multiple feature includes that the color of the decorative textile product is special
It seeks peace pattern characteristics；The sample image is smoothed, the smoothing processing be on the basis of gaussian filtering according to
Space length, color distance and neighborhood gradient distance adjust the sum weight of neighborhood territory pixel point to construct filter factor, then to institute
The each pixel for stating sample image carries out linear filtering；
Gray processing processing is carried out to the sample image after the smoothing processing using mean value method, and uses and automatically selects differentiation threshold
The method of value is split image using edge detection operator, to obtain the pattern characteristics of the sample image；Also pass through
The sample image of storage extracts rgb value, and the RGB value is converted to the HSV color space for being suitble to naked eyes to differentiate, and counts every
The pixel number of kind color component accounts for the ratio of image total pixel number, described to obtain the ratio distribution of image various colors component
Ratio is distributed for expressing color characteristic；
Textile color characteristic and pattern characteristics based on extraction count the frequency of homogenous characteristics appearance respectively, and are based on the frequency
It is secondary that the color characteristic and pattern characteristics are ranked up；
The part colours feature for sorting forward in the characteristic statistics unit and partial pattern feature is selected to carry out composition of geometry, and
Generate decorative textile product composition typesetting scheme preview original text.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811014178.6A CN109241390A (en) | 2018-08-31 | 2018-08-31 | A kind of intelligent design system and method for the decorative textile product based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811014178.6A CN109241390A (en) | 2018-08-31 | 2018-08-31 | A kind of intelligent design system and method for the decorative textile product based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109241390A true CN109241390A (en) | 2019-01-18 |
Family
ID=65059889
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811014178.6A Pending CN109241390A (en) | 2018-08-31 | 2018-08-31 | A kind of intelligent design system and method for the decorative textile product based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109241390A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112215793A (en) * | 2020-08-28 | 2021-01-12 | 阳信瑞鑫集团有限公司 | Method for drawing hand-embroidered carpet pattern detection texture and applying hand-embroidered carpet pattern detection texture to carpet weaving |
| CN112581432A (en) * | 2020-12-08 | 2021-03-30 | 中国纺织科学研究院有限公司 | Method and device for measuring mixing proportion of colored spun yarn, computer readable storage medium and electronic equipment |
| CN115482128A (en) * | 2022-08-18 | 2022-12-16 | 江苏牛掌柜科技有限公司 | Intelligent management method and system for hotel linen |
| CN117495986A (en) * | 2023-12-29 | 2024-02-02 | 宁波经纬数控股份有限公司 | Cloth pattern color acquisition method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104778703A (en) * | 2015-04-15 | 2015-07-15 | 天津大学 | Automatic extraction method for creative factors of fabric pattern |
| CN105844278A (en) * | 2016-04-15 | 2016-08-10 | 浙江理工大学 | Multi-feature fused fabric scanning pattern recognition method |
| CN106295654A (en) * | 2016-08-01 | 2017-01-04 | 深圳云天励飞技术有限公司 | A kind of image-recognizing method and device |
-
2018
- 2018-08-31 CN CN201811014178.6A patent/CN109241390A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104778703A (en) * | 2015-04-15 | 2015-07-15 | 天津大学 | Automatic extraction method for creative factors of fabric pattern |
| CN105844278A (en) * | 2016-04-15 | 2016-08-10 | 浙江理工大学 | Multi-feature fused fabric scanning pattern recognition method |
| CN106295654A (en) * | 2016-08-01 | 2017-01-04 | 深圳云天励飞技术有限公司 | A kind of image-recognizing method and device |
Non-Patent Citations (2)
| Title |
|---|
| 巩艳华 等: "基于颜色直方图的颜色特征提取", 《福建电脑》 * |
| 张诚: "纹理抑制平滑滤波及其在织物图案识别中的应用", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112215793A (en) * | 2020-08-28 | 2021-01-12 | 阳信瑞鑫集团有限公司 | Method for drawing hand-embroidered carpet pattern detection texture and applying hand-embroidered carpet pattern detection texture to carpet weaving |
| CN112215793B (en) * | 2020-08-28 | 2022-10-14 | 阳信瑞鑫集团有限公司 | Method for drawing hand-embroidered carpet pattern detection texture and applying hand-embroidered carpet pattern detection texture to carpet weaving |
| CN112581432A (en) * | 2020-12-08 | 2021-03-30 | 中国纺织科学研究院有限公司 | Method and device for measuring mixing proportion of colored spun yarn, computer readable storage medium and electronic equipment |
| CN115482128A (en) * | 2022-08-18 | 2022-12-16 | 江苏牛掌柜科技有限公司 | Intelligent management method and system for hotel linen |
| CN117495986A (en) * | 2023-12-29 | 2024-02-02 | 宁波经纬数控股份有限公司 | Cloth pattern color acquisition method and system |
| CN117495986B (en) * | 2023-12-29 | 2024-04-16 | 宁波经纬数控股份有限公司 | Cloth pattern color acquisition method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109241390A (en) | A kind of intelligent design system and method for the decorative textile product based on big data | |
| CN106055893B (en) | Garment coordination scheme generation method based on fashion template library and Auto-matching | |
| CN102360490B (en) | Color conversion and editing propagation-based method for enhancing seasonal feature of image | |
| CN105046280B (en) | A kind of wardrobe intelligent management apapratus and method | |
| KR102234609B1 (en) | Apparatus and method for recommending goods based on analysis of image database | |
| CN108776939A (en) | The analysis method and system of user power utilization behavior | |
| US6192150B1 (en) | Invariant texture matching method for image retrieval | |
| Fauqueur et al. | Region-based image retrieval: Fast coarse segmentation and fine color description | |
| CN105550660B (en) | A kind of recognition methods of woven fabric structure structure species | |
| CN106934386B (en) | A kind of natural scene character detecting method and system based on from heuristic strategies | |
| CN106250431B (en) | A kind of Color Feature Extraction Method and costume retrieval system based on classification clothes | |
| CN107833224B (en) | A kind of image partition method based on the synthesis of multilayer sub-region | |
| CN105023027B (en) | Live soles spoor decorative pattern image search method based on multiple feedback mechanism | |
| CN101806583A (en) | Microscopic image-based fiber fineness measurement method | |
| CN108537239A (en) | A kind of method of saliency target detection | |
| CN102637236B (en) | Intelligent color matching method for textile patterns | |
| CN105957124A (en) | Method and device for color editing of natural image with repetitive scene elements | |
| CN110276360A (en) | Generate computer installation, equipment, storage medium and the method for clothing matching scheme | |
| JP5337673B2 (en) | Image search system | |
| CN109636809A (en) | A kind of image segmentation hierarchy selection method based on scale perception | |
| CN102938053A (en) | Sugarcane characteristic extraction and recognition method based on computer vision | |
| CN103344571B (en) | A kind of marketable securities patience quality evaluating method and device | |
| Khotanzad et al. | Color image retrieval using multispectral random field texture model and color content features | |
| CN104239551B (en) | Multi-feature VP-tree index-based remote sensing image retrieval method and multi-feature VP-tree index-based remote sensing image retrieval device | |
| CN110110129A (en) | A kind of textile image search method based on fine textures characterization |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190118 |