CN110348530A - The method of lipstick color number for identification - Google Patents
The method of lipstick color number for identification Download PDFInfo
- Publication number
- CN110348530A CN110348530A CN201910643980.XA CN201910643980A CN110348530A CN 110348530 A CN110348530 A CN 110348530A CN 201910643980 A CN201910643980 A CN 201910643980A CN 110348530 A CN110348530 A CN 110348530A
- Authority
- CN
- China
- Prior art keywords
- picture
- lipstick
- trained
- identified
- color number
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of methods of lipstick color number for identification comprising the steps of: obtains training picture；Training picture is pre-processed；By pretreated trained picture training decision forest, decision forest includes multiple decision trees；Pass through pretreated trained picture training SVM classifier；Obtain picture to be identified；Picture to be identified is pre-processed；Pretreated picture to be identified is input to trained SVM classifier and decision forest obtains judging result；The judging result of SVM classifier and decision forest is voted to obtain final recognition result.The invention has the beneficial effects that picture is carried out hsv conversion by the method for the color number of lipstick for identification provided, eliminate the interference of shape and brightness to recognition result, and comprehensive descision is carried out using SVM classifier and decision forest simultaneously, improve the accuracy of lipstick color number identification.
Description
Technical field
The present invention relates to a kind of methods of lipstick color number for identification.
Background technique
Existing lipstick color classification method is mostly to utilize deep learning, the model counterpart of one convolutional neural networks of training
The drawbacks of red number is identified, existing method is: it is shape feature by the feature that the method for convolution is extracted, and lipstick
What classification relied on is color characteristic, and the result classified according to shape is very inaccurate, and uses the neural network of deep learning
The number of plies is generally too deep, and minimum for difference and lazy weight data are easy to over-fitting, has both identified to training data very quasi-
Really, untrained data are very inaccurate, and if the not deep picture resolution capability small for difference of the network number of plies is extremely weak.
Summary of the invention
The present invention provides a kind of methods of lipstick color number for identification, adopt the following technical scheme that:
A method of lipstick color number for identification comprising the steps of:
Obtain training picture；
Training picture is pre-processed；
By pretreated trained picture training decision forest, decision forest includes multiple decision trees；
Pass through pretreated trained picture training SVM classifier；
Obtain picture to be identified；
Picture to be identified is pre-processed；
Pretreated picture to be identified is input to trained SVM classifier and decision forest obtains judging result；
The judging result of SVM classifier and decision forest is voted to obtain final recognition result.
Further, training picture is carried out pretreated method particularly includes:
Noise reduction process is carried out to training picture；
Training picture after noise reduction process is subjected to hsv conversion；
It identifies and extracts the lipstick lotion region in the training picture after conversion.
Further, pretreated specific side is carried out to training picture are as follows:
Bilateral filtering removal mutation and noise are used to training picture；
Training picture after noise reduction process is subjected to hsv conversion；
It sets the threshold value identification of h and extracts the lipstick lotion region in training picture.
Further, pass through pretreated trained picture training decision forest method particularly includes:
Lipstick lotion region input decision forest in training picture after the conversion extracted is trained it.
Further, decision forest includes 3-6 decision tree.
Further, pass through pretreated trained picture training SVM classifier method particularly includes:
Count the distribution results of the h value of all the points in the lipstick lotion region in training picture；
Distribution results input SVM classifier is trained it.
Further, distribution results are statistic histogram；
The abscissa of statistic histogram be h value and ordinate be corresponding h value point number；
Wherein, the range of the precision of abscissa is more than or equal to 4 and to be less than or equal to 6.
Further, training picture is obtained method particularly includes:
It will be used for before trained lipstick is placed on white background；
The first image of the lipstick for training is obtained by RGB camera；
First image is stamped corresponding color labelled notation to obtain training picture；
Obtain picture to be identified method particularly includes:
Be analysed to color number lipstick be placed on white background before；
The second image of the lipstick of color number to be analyzed is obtained as picture to be identified by RGB camera.
Further, picture to be identified is carried out pretreated method particularly includes:
Noise reduction process is carried out to picture to be identified；
Picture to be identified after noise reduction process is subjected to hsv conversion；
It identifies and extracts the lipstick lotion region in the picture to be identified after conversion.
Further, picture to be identified is carried out pretreated method particularly includes:
Bilateral filtering removal mutation and noise are used to picture to be identified；
Picture to be identified after noise reduction process is subjected to hsv conversion；
The threshold value of setting h identifies and extracts the lipstick lotion region in picture to be identified.
The invention has the beneficial effects that picture is carried out hsv conversion by the method for the color number of lipstick for identification provided, go
Interference in addition to shape and brightness to recognition result, and comprehensive descision is carried out using SVM classifier and decision forest simultaneously, it improves
The accuracy of lipstick color number identification.
Detailed description of the invention
Fig. 1 is the flow chart of the method for the color number of lipstick for identification of the invention.
Specific embodiment
Specific introduce is made to the present invention below in conjunction with the drawings and specific embodiments.
As shown in Figure 1 for a kind of method of lipstick color number for identification of the invention, the color number for automatic identification lipstick.
The method of lipstick color number comprises the steps of: S1 for identification: obtaining training picture.S2: training picture is pre-processed.
S3: by pretreated trained picture training decision forest, decision forest includes multiple decision trees.S4: after pretreatment
Training picture training SVM (Support Vector Machine, support vector machines) classifier.S5: picture to be identified is obtained.
S6: picture to be identified is pre-processed.S7: by pretreated picture to be identified be input to trained SVM classifier and
Decision forest obtains judging result.S8: the judging result of SVM classifier and decision forest is voted to obtain final identification knot
Fruit.
For S1: obtaining training picture.
Firstly the need of obtain a certain number of trained pictures in subsequent step SVM classifier and decision forest instruct
Practice.For the ease of handling training picture, it will be used for before trained lipstick is placed on white background, obtained by RGB camera
The first image in trained lipstick is taken, then the color labelled notation that first image stamps corresponding lipstick is obtained to train figure
Piece.Background color can be selected according to the actual situation.
For S2: being pre-processed to training picture.
After getting trained picture by step S1, then training picture is pre-processed, it is pretreated method particularly includes:
Noise reduction process is carried out to training picture, the training picture after noise reduction process is subjected to hsv conversion, after identifying and extracting conversion
Lipstick lotion region in training picture.
Specifically, using bilateral filtering removal mutation and noise to training picture.The kernel function of bilateral filtering is empty
Between domain core and pixel coverage domain core synthesis result, in the flat site of image, pixel value is varied less, corresponding pixel coverage
Domain weight is close to 1, and spatial domain weight plays a major role at this time, is equivalent to carry out Gaussian Blur, in the fringe region of image, as
Element value variation is very big, and pixel coverage domain weight becomes larger, to maintain the information at edge.
Further, the training picture got is rgb picture, and the value that each pair of point is answered is (R, G, B), and HSV (Hue,
Saturation, Value) it is a kind of color space created according to the intuitive nature of color, also referred to as hexagonal pyramid model, this
The parameter of color is respectively in model: tone (H), saturation degree (S) and lightness (V), and the value of each point is (h, s, v) in picture.
Rgb image is subjected to hsv conversion method particularly includes: first will training picture each point the normalization of rgb value, i.e., by range from
0-255 zooms to 0-1.For each pixel, maximum value in rgb is taken to be calculated as max, i.e. max=max (R, G, B), minimum value is calculated as
Min, i.e. min=min (R, G, B), then each value is respectively as follows: in (h, s, v)
V=max (R, G, B)；
S=0 (if max=0), s=(max-min)/max (if max is not equal to 0)；
H=0 (if max=min), h=60* (g-b)/(max-min) (if max=r and g >=b), h=60*
(g-b)/(max-min)+360 (if max=r and g <b), h=60* (b-r)/(max-min)+120 (if max=g),
H=60* (r-g)/(max-min)+240 (if max=b)；
According to the above conversion relation, training picture is subjected to hsv conversion.To the threshold for the setting of the training picture after the changing h that walks around
Lipstick lotion region in value identification and extraction training picture.All points for meeting threshold value are specially picked out in full figure, it is right
These points carry out the processing of findcontours (searching profile), find maximum profile, contoured interior is as lipstick lotion area
Domain.
For S3: by pretreated trained picture training decision forest, decision forest includes multiple decision trees.
The lipstick lotion region in the training picture after the conversion extracted will be inputted into decision forest to it in step S2
It is trained.
Specifically, extracting the hsv feature of all the points on lotion, it regard each point as a data, every data is
The feature of 3 dimensions of hsv value, it is assumed that each color has 1000 points, then the training data of the color is 1000, due to every
The location of a point and shooting angle are different, therefore the value with the hsv of the point of of the same colour number lipstick lotion different location has carefully
Small deviation can obtain have the decision tree of color classification ability, decision so all the points using each color are trained
The determining device that tree is one 5 layers, each layer classify to input feature vector, finally obtain final result, the multiple decisions of training
Tree composition decision forest, is typically chosen 3-6 decision tree composition decision forest, and preferably 5 decision trees form decision forest.
For S4: passing through pretreated trained picture training SVM classifier.
Pass through pretreated trained picture training SVM classifier method particularly includes: count the lipstick in training picture
The distribution results of the h value of all the points in lotion region.Distribution results input SVM classifier is trained it.
Further, distribution results are statistic histogram.The abscissa of statistic histogram is the value of h, and ordinate is corresponding h
The number of the point of value.Wherein, abscissa precision is defaulted as 1, too sensitive to minor change, causes error too big, therefore, at this
In embodiment, it is set as the range of the precision of abscissa to be more than or equal to 4 and is less than or equal to 6, preferably 5.Such as 1-5,6-10 is
It is segmented one by one, counts the number of point of each segmentation as characteristic value.
For S5: obtaining picture to be identified.
Be analysed to color number lipstick be placed on white background before, the lipstick of color number to be analyzed is obtained by RGB camera
Second image can be selected according to the actual situation as picture to be identified, background color.
For S6: being pre-processed to picture to be identified.
Pretreated method is carried out to picture to be identified and to training the pretreated method of picture progress essentially identical, is treated
It handles picture and carries out noise reduction process, the picture to be processed after noise reduction process is subjected to hsv conversion, after identifying and extracting conversion
Lipstick lotion region in picture to be processed.
Specifically, using bilateral filtering removal mutation and noise to picture to be processed.The kernel function of bilateral filtering is
The synthesis result of spatial domain core and pixel coverage domain core, in the flat site of image, pixel value is varied less, corresponding pixel model
Domain weight is enclosed close to 1, spatial domain weight plays a major role at this time, it is equivalent to carry out Gaussian Blur, in the fringe region of image,
Pixel value changes very greatly, and pixel coverage domain weight becomes larger, to maintain the information at edge.Picture to be processed is carried out hsv to turn
It changes, identifies and extract the lipstick lotion region in picture to be processed to the threshold value for the setting of the picture to be processed after the changing h that walks around.Specifically
To pick out all points for meeting threshold value in full figure, these are put with the processing for carrying out findcontours (searching profile), is looked for
To maximum profile, contoured interior is as lipstick lotion region.
For S7: pretreated picture to be identified being input to trained SVM classifier and decision forest is sentenced
Disconnected result.
Pretreated picture to be identified is input to trained SVM classifier and decision forest, SVM classifier and certainly
Plan forest exports result respectively, wherein decision forest output result number with it includes the number of decision tree it is identical.
For S8: being voted the judging result of SVM classifier and decision forest to obtain final recognition result.
It to SVM classifier and determines after SVM classifier and decision forest export judging result, then through the method for ballot
The judging result of plan forest is voted, the last final recognition result of highest result of voting.
It, can point-device color number for identifying lipstick by all of above step.
The basic principles, main features and advantages of the invention have been shown and described above.The technical staff of the industry should
Understand, the above embodiments do not limit the invention in any form, all obtained by the way of equivalent substitution or equivalent transformation
Technical solution is fallen within the scope of protection of the present invention.
Claims (10)
1. a kind of method of lipstick color number for identification, which is characterized in that comprise the steps of:
Obtain training picture；
The trained picture is pre-processed；
By the pretreated trained picture training decision forest, the decision forest includes multiple decision trees；
Pass through the pretreated trained picture training SVM classifier；
Obtain picture to be identified；
The picture to be identified is pre-processed；
The pretreated picture to be identified is input to the trained SVM classifier and the decision forest is sentenced
Disconnected result；
The judging result of the SVM classifier and the decision forest is voted to obtain final recognition result.
2. the method for the color number of lipstick for identification according to claim 1, which is characterized in that
The trained picture is carried out pretreated method particularly includes:
Noise reduction process is carried out to the trained picture；
The trained picture after noise reduction process is subjected to hsv conversion；
It identifies and extracts the lipstick lotion region in the trained picture after conversion.
3. the method for the color number of lipstick for identification according to claim 2, which is characterized in that
Pretreated specific side is carried out to the trained picture are as follows:
Bilateral filtering removal mutation and noise are used to the trained picture；
The trained picture after noise reduction process is subjected to hsv conversion；
The threshold value of setting h identifies and extracts the lipstick lotion region in the trained picture.
4. the method for the color number of lipstick for identification according to claim 3, which is characterized in that
Pass through the pretreated trained picture training decision forest method particularly includes:
Lipstick lotion region in the trained picture after the conversion extracted is inputted the decision forest to instruct it
Practice.
5. the method for the color number of lipstick for identification according to claim 4, which is characterized in that
The decision forest includes 3-6 decision tree.
6. the method for the color number of lipstick for identification according to claim 3, which is characterized in that
Pass through the pretreated trained picture training SVM classifier method particularly includes:
Count the distribution results of the h value of all the points in the lipstick lotion region in the trained picture；
Distribution results input SVM classifier is trained it.
7. the method for the color number of lipstick for identification according to claim 6, which is characterized in that
The distribution results are statistic histogram；
The abscissa of the statistic histogram be h value and ordinate be corresponding h value point number；
Wherein, the range of the precision of abscissa is more than or equal to 4 and to be less than or equal to 6.
8. the method for the color number of lipstick for identification according to claim 1, which is characterized in that
Obtain the trained picture method particularly includes:
It will be used for before trained lipstick is placed on white background；
The first image of the lipstick for training is obtained by RGB camera；
The first image is stamped into corresponding color labelled notation and obtains the trained picture；
Obtain the picture to be identified method particularly includes:
Be analysed to color number lipstick be placed on white background before；
The second image of the lipstick of color number to be analyzed is obtained as the picture to be identified by RGB camera.
9. the method for the color number of lipstick for identification according to claim 1, which is characterized in that
The picture to be identified is carried out pretreated method particularly includes:
Noise reduction process is carried out to the picture to be identified；
The picture to be identified after noise reduction process is subjected to hsv conversion；
It identifies and extracts the lipstick lotion region in the picture to be identified after conversion.
10. the method for the color number of lipstick for identification according to claim 9, which is characterized in that
The picture to be identified is carried out pretreated method particularly includes:
Bilateral filtering removal mutation and noise are used to the picture to be identified；
The picture to be identified after noise reduction process is subjected to hsv conversion；
The threshold value of setting h identifies and extracts the lipstick lotion region in the picture to be identified.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910643980.XA CN110348530B (en) | 2019-07-17 | 2019-07-17 | Method for identifying lipstick number |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910643980.XA CN110348530B (en) | 2019-07-17 | 2019-07-17 | Method for identifying lipstick number |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110348530A true CN110348530A (en) | 2019-10-18 |
| CN110348530B CN110348530B (en) | 2021-11-12 |
Family
ID=68176591
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910643980.XA Active CN110348530B (en) | 2019-07-17 | 2019-07-17 | Method for identifying lipstick number |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110348530B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114818883A (en) * | 2022-04-07 | 2022-07-29 | 中国民用航空飞行学院 | CART decision tree fire image identification method based on optimal combination of color features |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030065525A1 (en) * | 2001-10-01 | 2003-04-03 | Daniella Giacchetti | Systems and methods for providing beauty guidance |
| CN101324955A (en) * | 2008-06-02 | 2008-12-17 | 昆明理工大学 | Method for dividing plant root system image based on color characteristic |
| CN104063722A (en) * | 2014-07-15 | 2014-09-24 | 国家电网公司 | Safety helmet identification method integrating HOG human body target detection and SVM classifier |
| CN105654469A (en) * | 2015-12-22 | 2016-06-08 | 深圳贝申医疗技术有限公司 | Infant stool color automatic analysis method and system |
| CN105787508A (en) * | 2016-02-22 | 2016-07-20 | 青岛大学 | Textile color identification method and system |
| CN109344869A (en) * | 2018-08-28 | 2019-02-15 | 东软集团股份有限公司 | A kind of disaggregated model optimization method, device and storage equipment, program product |
| CN109383168A (en) * | 2018-10-10 | 2019-02-26 | 上海市格致中学 | A kind of lipstick color number identification examination coloured pencil |
| CN109934092A (en) * | 2019-01-18 | 2019-06-25 | 深圳壹账通智能科技有限公司 | Identify color method, apparatus, computer equipment and storage medium |
| CN109948548A (en) * | 2019-03-20 | 2019-06-28 | 齐鲁工业大学 | A kind of the lipstick recommended method and system of the match colors based on machine learning |
-
2019
- 2019-07-17 CN CN201910643980.XA patent/CN110348530B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20030065525A1 (en) * | 2001-10-01 | 2003-04-03 | Daniella Giacchetti | Systems and methods for providing beauty guidance |
| CN101324955A (en) * | 2008-06-02 | 2008-12-17 | 昆明理工大学 | Method for dividing plant root system image based on color characteristic |
| CN104063722A (en) * | 2014-07-15 | 2014-09-24 | 国家电网公司 | Safety helmet identification method integrating HOG human body target detection and SVM classifier |
| CN105654469A (en) * | 2015-12-22 | 2016-06-08 | 深圳贝申医疗技术有限公司 | Infant stool color automatic analysis method and system |
| CN105787508A (en) * | 2016-02-22 | 2016-07-20 | 青岛大学 | Textile color identification method and system |
| CN109344869A (en) * | 2018-08-28 | 2019-02-15 | 东软集团股份有限公司 | A kind of disaggregated model optimization method, device and storage equipment, program product |
| CN109383168A (en) * | 2018-10-10 | 2019-02-26 | 上海市格致中学 | A kind of lipstick color number identification examination coloured pencil |
| CN109934092A (en) * | 2019-01-18 | 2019-06-25 | 深圳壹账通智能科技有限公司 | Identify color method, apparatus, computer equipment and storage medium |
| CN109948548A (en) * | 2019-03-20 | 2019-06-28 | 齐鲁工业大学 | A kind of the lipstick recommended method and system of the match colors based on machine learning |
Non-Patent Citations (4)
| Title |
|---|
| 亓呈明 等: "《机器学习、智能计算与高光谱遥感影像分类应用研究》", 31 May 2018, 中国财富出版社 * |
| 杨杰 等: "《视频目标检测和跟踪及其应用》", 31 August 2012, 上海交通大学出版社 * |
| 谢广明 等: "《自主视觉机器鱼竞赛》", 31 July 2013, 哈尔滨工程大学出版社 * |
| 赵池航 等: "《交通信息感知理论与方法》", 30 September 2014, 东南大学出版社 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114818883A (en) * | 2022-04-07 | 2022-07-29 | 中国民用航空飞行学院 | CART decision tree fire image identification method based on optimal combination of color features |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110348530B (en) | 2021-11-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Waghmare et al. | Detection and classification of diseases of grape plant using opposite colour local binary pattern feature and machine learning for automated decision support system | |
| Raut et al. | Plant disease detection in image processing using MATLAB | |
| Mebatsion et al. | Automatic classification of non-touching cereal grains in digital images using limited morphological and color features | |
| CN102117413B (en) | Method for automatically filtering defective image based on multilayer feature | |
| CN103634680B (en) | The control method for playing back and device of a kind of intelligent television | |
| CN109002851B (en) | Fruit classification method based on image multi-feature fusion and application | |
| CN109829914A (en) | The method and apparatus of testing product defect | |
| Almogdady et al. | A flower recognition system based on image processing and neural networks | |
| Liu et al. | Detection of citrus fruit and tree trunks in natural environments using a multi-elliptical boundary model | |
| CN108319973A (en) | Detection method for citrus fruits on tree | |
| CN109255344A (en) | A kind of digital display instrument positioning and Recognition of Reading method based on machine vision | |
| CN106778786A (en) | Apple disease recognition methods based on log-spectral domain laminated gradient direction histogram | |
| Saha et al. | Orange fruit disease classification using deep learning approach | |
| CN105574514A (en) | Greenhouse immature tomato automatic identification method | |
| CN110598574A (en) | Intelligent face monitoring and identifying method and system | |
| CN109284759A (en) | One kind being based on the magic square color identification method of support vector machines (svm) | |
| Lee et al. | Intelligent classification methods of grain kernels using computer vision analysis | |
| CN110348530A (en) | The method of lipstick color number for identification | |
| Yusuf et al. | Human face detection using skin color segmentation and watershed algorithm | |
| Nawawi et al. | Comprehensive pineapple segmentation techniques with intelligent convolutional neural network | |
| CN112115824B (en) | Fruit and vegetable detection method, fruit and vegetable detection device, electronic equipment and computer readable medium | |
| Maulana et al. | Color space and color channel selection on image segmentation of food images | |
| Subban et al. | Human skin segmentation in color images using gaussian color model | |
| Tamilselvi et al. | Color based K-Means Clustering For Image Segmentation to Identify the Infected Leaves | |
| Fitter | Strawberry detection under various harvestation stages |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |