AUTOMATED SKIN DISEASE DETECTION
PROJECT MEMBERS
MUHAMMAD ADNAN EJAZ 2111-FBAS-BSSE-F13
WAQAR YOUNAS KHAN 2112-FBAS-BSSE-F13
SUPERVISOR
MR. SYED MUHAMMAD SAQLAIN
DEPARTMENT OF COMPUTER SCIENCE AND SOFTWARE ENGINEERING
FACULTY OF BASIC AND APPLIED SCIENCES
INTERNATIONAL ISLAMIC UNIVERSITY ISLAMABAD
FINAL APPROVAL
Dated: ________________
It is certified that we have read the project report titled “(Automated Skin Diseases Detection)”
submitted by Muhammad Adnan Ejaz (2111/FBAS/BSSE/F13) and Waqar Younas Khan
(2112/FBAS/BSSE/F13). It is our judgment that this project is of sufficient standard to warrant
its acceptance by the International Islamic University, Islamabad for Bachelor’s Degree in
Software Engineering.
COMMITTEE
External Examiner:
Dr. Imran Khan ________________
Assistant Professor
Department of Computer Sciences & Software Engineering
International Islamic University,
Islamabad
Internal Examiner:
Dr. Husnain Abbass Naqvi ________________
Assistant Professor
Department of Computer Sciences & Software Engineering
International Islamic University, Islamabad
Supervisor:
Mr. Syed Muhammad Saqlain ________________
Assistant Professor
Department of Computer Sciences & Software Engineering
International Islamic University, Islamabad
"In the Name of ALLAH, the Most Beneficent, the Most Merciful"
Automated Skin Disease Detection Dissertation
IV
A dissertation submitted to
Department Of Computer Science & Software Engineering,
International Islamic University, Islamabad
As partial fulfillment of the requirements
For award of the degree of
Bachelors in Software Engineering.
Automated Skin Disease Detection Dedication
V
DEDICATION This project is dedicated to our parents, brothers, sisters, family members, teachers and
especially Mr. Syed Muhammad Saqlain who have been a great source of motivation,
inspiration and supported us all the way since the beginning of our studies and project.
Automated Skin Disease Detection Declaration
VI
DECLARATION
We hereby declare that we developed this software and this report entirely on the basis of our
personal efforts made under the sincere guidance of our project supervisor Mr. Syed
Muhammad Saqlain. No portion of this work presented in this report has been submitted in
support of our applications for any other degree or qualification of this or any other University
or institute of learning. We further declare that this software and all associated documents,
reports are submitted as partial requirements for the degree of Bachelor of Science in Software
Engineering.
Muhammad Adnan Ejaz
2111/FBAS/BSSE/F13
Waqar Younas Khan
2112/FBAS/BSSE/F13
Automated Skin Disease Detection Acknowledgement
VII
ACKNOWLEDGEMENT
We humbly thank ALLAH (S.W.T) Almighty, the Merciful and the Beneficent, Who gave us
health, thoughts and co-operative people to enable us achieve this goal.
We would like to record our gratitude to our supervisor, Mr. Syed Muhammad Saqlain for his
supervision, advice, and guidance to develop an understanding of the project. Above all and the
most needed, he provided us unflinching encouragement and support in various ways.
Finally, we would like to thank everybody who was important to the successful realization of
project including our Parents, brothers, sisters, family members, teachers and friends (Hamza
Javed, Jibran Ahmed Siddiqui, Awais Mushtaq and others).
Automated Skin Disease Detection Project in Brief
VIII
PROJECT IN BRIEF
Project Title Automated Skin Disease Detection
Version 1.0
Core Team Muhammad Adnan Ejaz
Waqar Younas Khan
Supervised by Mr. Syed Muhammad Saqlain
Date Started May 2017
Date Completed October 2017
Language/Technology C#, Aforge Library, Image Processing
System Used Core-2duo, RAM-2GB, Microsoft windows
8.1
Automated Skin Disease Detection Abstract
IX
ABSTRACT
This project (Automated Skin Diseases Detection) is developed in C# technology using image
processing Techniques. In this project, an approach for automated segmentation and
classification of skin diseases is proposed. Initially, coloured skin images are filtered and
converted to grayscale images to remove unwanted hairs and noise. Then the segmentation
process is carried out to extract disease areas. For segmentation, K-means clustering method
is applied. From the disease image, we extracted color features such as; Mean, Standard
deviation, Skewness, Variance and Kurtosis with the help of RGB Histogram. SVM classifier
is used for the classification.
Automated Skin Disease Detection Table of Contents
X
Table of Contents
CHAPTER 1 ......................................................................................................................................................... 1
INTRODUCTION ................................................................................................................................................ 1
1. Introduction .................................................................................................................................................. 2
1.1. Existing Systems .................................................................................................................................. 2
1.2. Problem Statement .............................................................................................................................. 3
1.3. Proposed Solution ............................................................................................................................... 3
1.4. Features ................................................................................................................................................ 5
1.5. Tools & Technologies .......................................................................................................................... 5
CHAPTER 2 ......................................................................................................................................................... 6
SYSTEM ANALYSIS .......................................................................................................................................... 6
2. System Analysis ............................................................................................................................................ 7
2.1. Use Case Model ................................................................................................................................... 7
2.1.1. Use Case Diagram ........................................................................................................................... 8
2.1.2. Use Case Description ...................................................................................................................... 9
2.2. System Sequence Diagrams .............................................................................................................. 15
2.2.1. Browse Image .................................................................................................................................... 15
2.2.2. Gaussian Blur .................................................................................................................................... 16
2.2.3. Gray-scale Conversion ..................................................................................................................... 16
2.2.4. Segmentation ..................................................................................................................................... 17
2.2.5. Features Extraction .......................................................................................................................... 17
2.2.6. Classification ..................................................................................................................................... 18
2.3. Domain Model ................................................................................................................................... 19
2.4. Activity Diagram ............................................................................................................................... 20
CHAPTER 3 ....................................................................................................................................................... 21
SYSTEM DESIGN ............................................................................................................................................. 21
3. System Design ............................................................................................................................................. 22
3.1. Sequence Diagram ............................................................................................................................... 22
3.2. Class Diagram ...................................................................................................................................... 23
CHAPTER 4 ....................................................................................................................................................... 24
IMPLEMENTATION ........................................................................................................................................ 24
4. Implementation ........................................................................................................................................... 25
Automated Skin Disease Detection Table of Contents
XI
4.1 Gaussian Blur .................................................................................................................................... 25
4.2 K-means Clustering .......................................................................................................................... 25
4.3 RGB Color Features Extraction ...................................................................................................... 26
4.4 Classification ..................................................................................................................................... 26
4.4.1. SVM (Support Vector Machine) Algorithm ................................................................................... 27
4.4 Package Diagram .............................................................................................................................. 28
4.5 Deployment Diagram ........................................................................................................................ 28
CHAPTER 5 ....................................................................................................................................................... 29
TESTING ............................................................................................................................................................ 29
5. Test Cases .................................................................................................................................................... 30
5.1. Test Case TC-01: Browse Image ...................................................................................................... 30
5.2. Test Case TC-02: Image Pre-processing ......................................................................................... 31
5.3. Test Case TC-03: Segmentation ....................................................................................................... 32
5.4. Test Case TC-04: Features Extraction ............................................................................................ 33
5.5. Test Case TC-05: Classification ....................................................................................................... 34
CHAPTER 6 ....................................................................................................................................................... 35
CONCLUSION ................................................................................................................................................... 35
6. Conclusion ................................................................................................................................................... 36
CHAPTER 7 ....................................................................................................................................................... 37
USER MANUAL ................................................................................................................................................ 37
7. User Manual ................................................................................................................................................ 38
7.1. Browse Image .................................................................................................................................... 38
7.2. Pre-Processing ................................................................................................................................... 39
7.3. Segmentation & Features Extraction .............................................................................................. 39
7.4. Diagnosis (Classification) ................................................................................................................. 40
REFERENCES ................................................................................................................................................... 41
References ........................................................................................................................................................... 42
CHAPTER 1
INTRODUCTION
Chapter 01 Introduction
Automated Skin Disease Detection 02
1. Introduction
Skin being the largest organ/part of a human body, covers the all parts
of the body. Skin functionality in a human body is of more importance. A small disorder in its
functionality might affect the functionality of other parts of the human body. So, a proper care
is necessary for the skin as it is mostly exposed to the outer environment. Otherwise, in case
of any infection or disease it can also affect the other parts of the human body. In case of any
infection or disease, it should be dealt with proper care and treatment otherwise, these
infections and diseases can be worst. There are different types of skin diseases and infections
such as; chickenpox, melanoma, warts, shingles, leprosy and etc. That part of the skin that is
infected is called as skin lesion area. Inexperienced dermatologist might not be able to detect
and differentiate such kind of infections and diseases without any computer-aided diagnostic
system. So, that is the reason that computer aided systems are now an integral part of the
medical field.
In this project, we are working on the detection, segmentation and classification of the skin
diseases by applying C# and image processing techniques & algorithms. The proposed
methodology of the system is listed below as;
Browse Image
Pre-processing
Segmentation
Features Extraction
Classification
1.1. Existing Systems
There are some systems similar to it. But they cannot provide
such functionality in one application. Such as;
Melanoma Detection (Mobile application)
Chapter 01 Introduction
Automated Skin Disease Detection 03
1.2. Problem Statement
Dermatology is an important field of medical sciences that
deals with the skin, hair and nails diseases. But among of these, skin has a wide field of study
as there are many diseases that are associated with the skin. These diseases require a proper
treatment and care. Otherwise, these diseases can be turn into severe diseases that can cause
the death. Mole is found almost on the skin of every human being. Without proper care it can
be turn into melanoma that is considered as the skin cancer. There are many other skin diseases
that can turn into severe diseases without proper care. Some skin diseases evolve gradually and
some do not. There are some skin diseases that have no prior symptoms and signs. It is
necessary for a dermatologist to be aware of the different conditions and stages of the skin
diseases and that is difficult for those who are in-experienced. So, for this reason computer
aided systems were developed to help the dermatologist to diagnose different kind of skin
diseases using different techniques and algorithms of image processing. Each disease is
categorized and classified with different image processing techniques.
But majority of such systems are developed in Matlab, C++, Python and Java. While a little
bit of working has been done in other computer languages.
1.3. Proposed Solution
Automated Skin Disease Detection will assist the
dermatologist to diagnose the skin disease and the different stages of that disease. Primary
objective of this project is to develop a system that can detect an image of skin disease and
classify it by applying advance algorithms and techniques of C# and image processing.
Chapter 01 Introduction
Automated Skin Disease Detection 04
Figure 1.1 Overview of proposed System
Chapter 01 Introduction
Automated Skin Disease Detection 05
1.4. Features
Main features of the Automated Skin Disease Detection System are given
below.
Browsed image conversion into gray-scale image
Noise and hair removal from the gray-scale image
Segmentation by applying K-means clustering
RGB Histogram of the image
Color features extraction based on the RGB histogram
Classification of the skin disease
1.5. Tools & Technologies
Following tools and technologies are used to develop this
system.
Microsoft Visual Studio 2013
C# as a development language and Aforge C# Library
Image Processing algorithms and techniques
CHAPTER 2
SYSTEM ANALYSIS
Chapter 02 System Analysis
Automated Skin Disease Detection 07
2. System Analysis
It is a Software Engineering task that is used to reduce the gap
between software design & system level software engineering.
2.1. Use Case Model
Use Case Model Includes:
Use Case Diagram
Use Case Descriptions
System Sequence Diagrams
Chapter 02 System Analysis
Automated Skin Disease Detection 08
2.1.1. Use Case Diagram
Figure 2.1 Use Case Diagram
Chapter 02 System Analysis
Automated Skin Disease Detection 09
2.1.2. Use Case Description
In software and systems engineering, a use case is a list of
actions or event steps typically defining the interactions between a role (known in the Unified
Modeling Language as an actor) and a system to achieve a goal. The actor can be a human or
other external system.
2.1.2.1. Use Case UC 01: Browse Image
Use Case ID UC-01
Scope Automated Skin Disease Detection
Name Browse Image
Primary Actor User
Goal Browse & Select an Image
Pre-Condition There Should be an image in the target folder.
Post-Condition Image converted to specific size required by the app.
Image Browsed Successfully.
User can see the image uploaded in the application window.
Main Success Scenario 1: User click on the Browse Image Button.
2: User Will select image from the target folder.
3: System will load the image.
4: User can see the image uploaded in the application window.
Alternate If image can’t be uploaded, an error message will be prompted
Table 2.1 Browse Image
Chapter 02 System Analysis
Automated Skin Disease Detection 10
2.1.2.2. Use Case UC 02: Gaussian Blur
Use Case ID UC-02
Scope Automated Skin Disease Detection
Name Gaussian Blur
Primary Actor User
Goal To remove noise and un-wanted things from the image such
as hair.
To smooth the image for better result.
To enhance the image quality.
Pre-Condition Use-Case UC-01 is completed successfully.
Image is uploaded in the application window.
Post-Condition Noise & unwanted things removed successfully from the
image.
Image is smoothed, enhanced and clear.
Main Success Scenario There is an uploaded image in the application window.
Uploaded image enhanced and smoothed by applying
Gaussian blur filter.
Alternate If image can’t be smoothed and enhanced, an error message
will be prompted.
Table 2.2 Gaussian Blur
Chapter 02 System Analysis
Automated Skin Disease Detection 11
2.1.2.3. Use Case UC 03: Gray-Scale Conversion
Use Case ID UC-03
Scope Automated Skin Disease Detection
Name Gray-Scale Conversion
Primary Actor User
Goal To convert the smoothed color image into Grayscale.
Pre-Condition Use-Case UC-02 is completed successfully.
Post-Condition Smoothed image successfully converted into grayscale image.
User can see the grayscale image in the application window.
Main Success Scenario There is a smoothed image in the application window.
User click the “Convert to grayscale” button.
Application converted the image into grayscale.
User can see the converted grayscale image in the application
window.
Alternate If image can’t be converted to grayscale, then error message
will be prompted by the application.
Table 2.3 Gray Scale Conversion
Chapter 02 System Analysis
Automated Skin Disease Detection 12
2.1.2.4. Use Case UC 04: Segmentation
Use Case ID UC-04
Scope Automated skin Disease Detection
Name Segmentation
Primary Actor User
Goal To sub-divide an image into two colors region by applying 2
clusters.
Pre-Condition Use-Case UC-03 is completed successfully.
Pre-processing (Gaussian blur & Grayscale conversion) of the
image is completed successfully.
Post-Condition Segmentation of the grayscale image is completed
successfully.
Sub-division of the grayscale image into two colors is
completed successfully.
User can see the clustered image in the application window.
Main Success Scenario There is clustered (sub-divided) image in the application
window.
Image is sub-divided into two different colors.
User can see the segmented/clustered image in the application
window.
Alternate If image can’t be sub-divided/segmented/clustered, an error
message will prompt.
Table 2.4 Segmentation
Chapter 02 System Analysis
Automated Skin Disease Detection 13
2.1.2.5. Use Case UC 05: Features Extraction
Use Case ID UC-05
Scope Automated Skin Disease Detection
Name Features Extraction
Primary Actor User
Goal To find the color features (Mean, standard deviation,
variance, skewness and kurtosis) through RGB histogram.
Pre-Condition Use-Case UC-04 is completed successfully.
Segmented/clustered Grayscale image should be converted
to color image.
Post-Condition Color features extracted successfully and shown on the
application window.
A separated RGB histogram based on the extracted features
and color (RED, Blue & Green) is shown on the application
window.
Main Success Scenario There is an image with extracted color features.
Separate histogram of each color Red, Blue and Green can
be seen on the application window.
Alternate If image desired features can’t be extracted, an error message
will be prompted.
If there is grayscale image instead of color image, an error
message will be prompted.
Table 2.5 Features Extraction
Chapter 02 System Analysis
Automated Skin Disease Detection 14
2.1.2.6. Use Case UC 06: Classification
Use Case ID UC-06
Scope Automated Skin Disease Detection
Name Classification
Primary Actor User
Goal To classify the image based on the extracted color features of
the image.
Pre-Condition Use-Case UC-05 is completed successfully.
Image color features has been extracted successfully.
Post-Condition Predicted image based on the extracted features data
matching.
Image classified successfully.
Main Success Scenario There is an image with extracted color features.
Image matched and classified successfully based on the
features.
Alternate If Image can’t be classified due to mismatch of the features,
an error message will be prompted.
Table 2.6 Classification
Chapter 02 System Analysis
Automated Skin Disease Detection 15
2.2. System Sequence Diagrams
In software engineering, a system sequence
diagram (SSD) is a sequence diagram that shows, for a particular scenario of a use case, the
events that external actors generate, their order, and possible inter-system events. System
sequence diagrams are visual summaries of the individual use cases. All systems are treated as
a black box; the diagram places emphasis on events that cross the system boundary from actors
to systems. A system sequence diagram should be done for the main success scenario of the use
case, and frequent or complex alternative scenarios.
2.2.1. Browse Image
Figure 2.2 Browse Image
Chapter 02 System Analysis
Automated Skin Disease Detection 16
2.2.2. Gaussian Blur
Figure 2.3 Gaussian Blur
2.2.3. Gray-scale Conversion
Figure 2.4 Gray-scale Conversion
Chapter 02 System Analysis
Automated Skin Disease Detection 17
2.2.4. Segmentation
Figure 2.5 Segmentation
2.2.5. Features Extraction
Figure 2.6 Features Extraction
Chapter 02 System Analysis
Automated Skin Disease Detection 18
2.2.6. Classification
Figure 2.7 Classification
Chapter 02 System Analysis
Automated Skin Disease Detection 19
2.3. Domain Model
Figure 2.8 Domain Model
Chapter 02 System Analysis
Automated Skin Disease Detection 20
2.4. Activity Diagram
Figure 2.9 Activity Diagram
CHAPTER 3
SYSTEM DESIGN
Chapter 03 System Design
Automated Skin Disease Detection 22
3. System Design
System design is the process of defining the components, modules,
interfaces, and data for a system to satisfy specified requirements. System development is the
process of creating or altering systems, along with the processes, practices, models, and
methodologies used to develop them.
3.1. Sequence Diagram
Figure 3.1 Sequence Diagram
Chapter 03 System Design
Automated Skin Disease Detection 23
3.2. Class Diagram
Figure 3.2 Class Diagram
CHAPTER 4
IMPLEMENTATION
Chapter 04 Implementation
Automated Skin Disease Detection 25
4. Implementation
This chapter will elaborate the interface along with the basic
principles that are kept in view while designing the interface. An idea can become worthless
when it is not conveyed properly. This chapter introduces how this application is implemented.
4.1 Gaussian Blur
The main purpose of the image Pre-processing is to enhance and
restore the image. To remove the noise from the image, we used 3x3 Gaussian blur. It smooth
the image because the Gaussian smoothing in 2D convolution operation is used ‘blur’ images
and remove hair and noise.
Figure 4.1 Implementation of Gaussian blur
4.2 K-means Clustering
Here K-means Clustering is being used in Segmentation
process. As we know that Segmentation is used to subdivide the image into its constituent
objects or regions. K-means clustering work as per given number of clusters. If we want an
image to be in two cluster regions, then we will allocate two clusters before process. K-means
algorithm will act upon allocated number of clusters and will divide the image into two clusters
that will be different from each other on the basis of colour present in that image. K-means
algorithm will run until it completely subdivide the image into two clusters based on the colour
present in the image.
Chapter 04 Implementation
Automated Skin Disease Detection 26
Figure 4.2 Clustered colours after implementation of K-means Clustering
4.3 RGB Color Features Extraction
In RGB Color Features Extraction, we
extracted the five different statistics from the disease image based on the Red, Green and Blue
color of the image. These five statistics are “Mean, Standard Deviation, Skewness, Variance
and Kurtosis”. After extraction, these features are also shown in the application along with
their calculations.
4.4 Classification
Here in classification, we divide the datasets into training and testing
data. Then using this data in SVM (Support Vector Machine) Algorithm, we classify the
diseases whether it is a Melanoma, Naevus or just a Normal Mole.
Chapter 04 Implementation
Automated Skin Disease Detection 27
4.4.1. SVM (Support Vector Machine) Algorithm
SVM (Support Vector Machine)
algorithm is a machine learning algorithm that is mainly used for classification and regression
analysis. SVM build a model based on training and testing data. In SVM, classification is
performed by finding the hyper-plane that is further used to differentiate the two classes. Then
it calculates the maximize margin between the nearest data. SVM finally select that data in the
hyper plane in the process of classification that has no error.
Figure 4.3 SVM Process Overview
Figure 4.4 Working Illustration of the SVM
Chapter 04 Implementation
Automated Skin Disease Detection 28
4.4 Package Diagram
Figure 4.5 Package Diagram
4.5 Deployment Diagram
Figure 4.6 Deployment Diagram
CHAPTER 5
TESTING
Chapter 05 Testing
Automated Skin Disease Detection 30
5. Test Cases
A test case is a set of conditions or variables under which a tester will
determine whether a system under test satisfies requirements or works correctly.
5.1. Test Case TC-01: Browse Image
Test Case ID TC -01
Functional Area/Module Browsing Image
Purpose To check the Browsing/loading of an image to the
application.
Action to Perform 1. User click on Browse image button
2. Browsing window opens
3. User selects image
4. Application load selected image
Prerequisites Application is running
Test Case Engineer Waqar Younas Khan
Environment Windows 8.1
Expected Result(s) Image Browsed/loaded successfully
Comments: Test passed successfully
Table 5.1 TC-01- Browse Image
Chapter 05 Testing
Automated Skin Disease Detection 31
5.2. Test Case TC-02: Image Pre-processing
Test Case ID Test -02
Functional Area/Module Image Pre-processing
Purpose To smooth the image.
To remove noise from the image to enhance it.
Action to Perform 1. User click on Pre-processing button
2. Gaussian blur is applied
3. Gray-scale conversion
4. Resulted image is shown
Prerequisites Application is running
Test Case Engineer Muhammad Adnan Ejaz
Environment Windows 8.1
Expected Result(s) Pre-processing done successfully
Comments: Test passed successfully
Table 5.2 TC-02- Image Pre-processing
Chapter 05 Testing
Automated Skin Disease Detection 32
5.3. Test Case TC-03: Segmentation
Test Case ID Test -03
Functional Area/Module Segmentation
Purpose To subdivide the image into its constituent regions or
objects.
Action to Perform 1. User click on Segmentation button
2. Selected clusters are applied
3. Segmentation should stop when the objects of interest in
an application have been isolated.
Prerequisites Application is running
Test Case Engineer Waqar Younas Khan
Environment Windows 8.1
Expected Result(s) Segmentation done successfully
Comments: Test passed successfully
Table 5.3 TC-03- Segmentation
Chapter 05 Testing
Automated Skin Disease Detection 33
5.4. Test Case TC-04: Features Extraction
Test Case ID Test -04
Functional Area/Module Features Extraction
Purpose To check & extract features of the image in application.
Action to Perform 1. Start features extraction
2. Re-convert the gray-scale segmented image to colour
image
3. Extract RGB color features of the image
4. Map the color features histogram based on each colour
Red, Green and Blue of the image
Prerequisites Application is running
Test Case Engineer Muhammad Adnan Ejaz
Environment Windows 8.1
Expected Result(s) Features extracted successfully
Comments: Test passed successfully
Table 5.4 TC-04- Features Extraction
Chapter 05 Testing
Automated Skin Disease Detection 34
5.5. Test Case TC-05: Classification
Test Case ID Test -05
Functional Area/Module Classification
Purpose To check the application for classification.
Action to Perform 1. Start Classification
2. System train the classifier on training data.
3. System test the classifier over test data.
5. System calculate accuracy based on results obtained
through test data.
Prerequisites Application is running
Test Case Engineer Muhammad Adnan Ejaz
Waqar Younas Khan
Environment Windows 8.1
Expected Result(s) Classification done successfully
Comments: Test passed successfully
Table 5.5 TC-05- Classification
CHAPTER 6
CONCLUSION
Chapter 06 Conclusion
Automated Skin Disease Detection 36
6. Conclusion
We feel very proud after development of “Automated Skin Disease
Detection” application successfully. Before starting of this project we have the theoretical
knowledge of software engineering, but it is far away from theory to develop a real life system
that completely fulfill user requirements.
During the development of the project we learn about many things that are listed below:
Different kind of skin diseases with their types.
How a simple skin disease can be severe without proper care.
How a simple mole can turn into a cancer.
Different types of skin cancers and their causes.
Image processing and its role in medical sciences.
Project management
Latest tools and technologies
Testing strategies
CHAPTER 7
USER MANUAL
Chapter 07 User Manual
Automated Skin Disease Detection 38
7. User Manual
User Manual of any application help the user how to operate the
application. It provides the overview of the application.
7.1. Browse Image
Figure 7.1: Browse Image
Chapter 07 User Manual
Automated Skin Disease Detection 39
7.2. Pre-Processing
Figure 7.2: Pre-processing
7.3. Segmentation & Features Extraction
Figure 7.3: Segmentation & Features Extraction
Chapter 07 User Manual
Automated Skin Disease Detection 40
7.4. Diagnosis (Classification)
Figure 7.4: Diagnosis (Classification)
REFERENCES
References
Automated Skin Disease Detection 42
References
1. https://www.sciencedirect.com/science/article/pii/S1877050915003269
2. http://haishibai.blogspot.com/2009/09/image-processing-c-tutorial-4-gaussian.html
3. https://www.codeproject.com/Articles/33838/Image-Processing-using-C
4. https://code.msdn.microsoft.com/windowsapps/How-to-convert-color-image-7b16899d
5. https://github.com/accord-
net/framework/blob/master/Sources/Accord.MachineLearning/Clustering/KMeans/KMeans.cs#L293
6. https://www.codeproject.com/Questions/216582/Help-about-Image-Segmentation-with-K-means
7. https://www.daniweb.com/programming/software-development/threads/244974/k-means-clustering-
algorithm
8. http://www.codeding.com/articles/k-means-algorithm
9. https://www.codeproject.com/Articles/35895/Computer-Vision-Applications-with-C-Part-II
10. http://www.aforgenet.com/framework/features/image_statistics.html
11. https://github.com/ccerhan/LibSVMsharp
12. https://www.daniweb.com/programming/software-development/threads/373771/divide-dataset-into-
training-and-test-data-set-using-random-sampling
13. https://www.ranorex.com/forum/test-data-from-csv-file-t2395.html
14. https://www.deepdetect.com/tutorials/csv-training/
15. https://blog.testproject.io/2017/02/09/read-data-csv-file-in-c/
16. https://www.skincancer.org/skin-cancer-information/melanoma/the-stages-of-melanoma
17. http://www.dermnet.com/contacts.php
18. https://www.dermnetnz.org/image-catalogue/
19. https://www.mayoclinic.org/diseases-conditions/skin-cancer/multimedia/melanoma/sls-20076095?s=3
20. http://www.fc.up.pt/addi/ph2%20database.html
21. https://www.flickr.com/groups/2733406@N25/pool/
22. http://biogps.org/dataset/tag/melanoma/
23. https://www.melanoma.org/understand-melanoma/resource-library/pictures-of-melanoma