WPR3

Amity School of Engineering & Technology

Department of Computer Science and Engineering

WEEKLY PROGRESS REPORT (WPR)

    For the week :  19^th May 2014 to 26^th May 2014 

WPR 2

Program: B. Tech. CSE (2012-14)                 Enrollment Number: A2305212207 and A2305212211                                                              

Faculty Guide’s Name: Ms. Shilpi Gupta         Name: Karan Razdan and Raghav Gogia

                                                                                                                                                       

TITLE OF THE PROJECT:

Optical Charter Recognition

TARGETS FOR THE WEEK:

v  Study more about the image processing toolbox available in MATLAB .

v  Got a new Version of MATLAB(R2014b) and inspected the various new features offered by it.

v  Further enhanced our database by adding a new font New times roman.

v  Studied the Language tool Api documentation and how it can be implemented.

ACHIEVEMENTS FOR THE WEEK: 

      In the third week, we:

v We enhanced our pervious created data base of our OCR by adding a new font Times new roman to our previously created data base of Ariel.

v We studied the Language tool api for its various dictionaries and functions and tried to implement it on a small program which would check a input for any spelling mistakes.

v Brushed up our Java GUI development skills.(Applets, swings , layouts and various events).

v Revived the mat lab r2014a newly introduced function ocr. N how can we implement it to our needs.

FUTURE WORK PLANS:

v  Study the algorithm in greater depth and try to fully understand all the concepts involved.

v  Expand our database and make it more advanced.

v  Develop a gui for our project.

v  In coming week plan is to enhance the output of our previous week work and debug it for some errors.

Progress Report 2

Amity School of Engineering & Technology

DepartmentofComputerScienceandEngineering

WEEKLY PROGRESS REPORT (WPR)

For the week :12^th May 2014 to 18^th May 2014       

WPR 2

Program: B.Tech. CSE (2012-14)Enrollment Number: A2305212207and A2305212211

Faculty Guide’s Name: Ms. ShilpiGuptName:Karan Razdanand RaghavGogia

                                                                                                                                                   

TITLE OF THE PROJECT:

Optical Charter Recognition

TARGETS FOR THE WEEK:

v  Study more about the image processing toolbox available in MATLAB.

v  Start looking for algorithms to help in processing images and reading alphabets from it.

v  Start building a basic database for image processing.

v  Start searching for open source dictionaries which would help with the spell check feature.

ACHIEVEMENTS FOR THE WEEK:

      In the second week, we:

v Started the process of creating a basic database. This database would be used as a reference for the image being processed. Each letter will be recognized and matched from this database and hence be recognized. The more detailed and advanced the database, the more accurate the recognition is. For our initial starting database, the first decision to be made was about the font to be used. We decided upon one of the most widely used fonts, the ‘Ariel’ font. A simple database was creates based on this font system.

v Started looking for an open source Java API which would help us in the feature of spell checking. We arrived at 3 options; The Sentry Spell Checker, Jazzy and LanguageTool API. The first one was a commercial product and had to be bought, so we didn’t select it. The second one was an open source API which suited our needs but it was not under any current active development so we chose the third one, the LanguageTool API. After some research, we figured out it was the best option for us. The LaguageTool API provides us with spell checking capabilities and at the same time it makes it easy to add more powerful checks such as several grammatical checks.

v Started brushing up on our Java programming concepts.

v Came across the Genetic Algorithm. It can be used to increase the efficiency of the database and make the overall process much more accurate. But the concepts involved need some more studying in order to fully understand the algorithm and be able to use it to our advantage.

FUTURE WORK PLANS:

v  Study the algorithm in greater depth and try to fully understand all the concepts involved.

v  Expand our database and make it more advanced.

v  Further study the LanguageTool API.

Weekly Progress Report ONE (Updated)

WPR of Week 1:  5^th May 2014 to 11^th May 2014   

program: B. Tech. CSE (2012-14)                           

Faculty Guide’s Name: Ms. Shilpi Gupta

                                                                                                                                                                                                                                    

Name: Raghav Gogia and Karan Razdhan

Enrollment Number: A2305212211  and A2305212207

TITLE OF THE PROJECT:

Optical Charter Recognition

TARGETS FOR THE WEEK:

To learn the basic concepts of image processing and various algorithms available for the development of our project, and getting familiar with the tools required to create an application

Before starting the project we need a  language or a tool so we choose  MATLAB coz of following advantages offered by MATLAB

•  A very large (and growing) database of built-in algorithms for image processing and computer vision applications
• MATLAB allows you to test algorithms immediately without recompilation. You can type something at the command line or execute a section in the editor and immediately see the results, greatly facilitating algorithm development.
• The MATLAB Desktop environment, which allows you to work interactively with your data, helps you to keep track of files and variables, and simplifies common programming/debugging tasks
• The ability to read in a wide variety of both common and domain-specific image formats.
• A large user community with lots of free code and knowledge sharing
• The ability to auto-generate C code, using MATLAB Coder, for a large (and growing) subset of image processing and mathematical functions, which you could then use in other environments, such as embedded systems or as a component in other software.

ACHIEVEMENTS FOR THE WEEK: 

In first week 1 studied about

v  What MATLAB is all about

v  Application Of MATLAB

v  Different Toolbox available in MATLAB

v  Studied the Image Processing Tool Box Offered by MATLAB

v  Also went through different resources available at our disposal for the development and studied earlier available papers.

v  Major achievement of the week was the Starting of the open source platform(BLOG:- http://summerprojectamity.blogspot.in/) for the publishing our research on a larger scale and not restricting our research and knowledge to our self rather den publishing it .On weekly bases.

FUTURE WORK PLANS:

v  In the upcoming week we are looking forward for studding different algorithms for processing a image and reading patterns from it(alphabets ).

v   Looking for open source dictionaries available for the spell check feature

v   Development of a GUI for our OCR .

.

WEEK’S SUMMARY:

Days	Progress MADE
Monday (5-May)	Studied Basis of MATLAB (Like writing a simple mat lab script , input and output using MATLAB at run time ,creating and opening a text file)
Tuesday (6-May)	Studied Application Of MATLAB and different toolboxes offered by MATLAB (such as Simulink with a core focus of how on which all platforms MATLAB can work and with which all languages matlab code can be executed(working of  matlab with java was our core research topic))
Wednesday (7-May) and Thursday    (8-may)	Studied the Image Processing Tool Box Offered by MATLAB Studied various Functions for our utility offered by dis tool box like. Few imp of them our Imread- for loading image into memory Imshow for displaying the image Imtool for viewing the image and processing it interactively. Thresholding of image to grayscale Eg threshold = graythresh(image); image =~im2bw(image,threshold);
Friday (9-May)	Went through different resources available at our disposal for the development and studied earlier available papers.
Saturday (10-May)	Searched for all the open source platforms where we could use to share our research with world. And weighting the advantages and disadvantages of each platform and selecting BlogSpot by google as a final decision. Because of its several feature like ease of writing and easily editing and support for multiple editors and and contributors.
Sunday (11-May)	Created our Blog and published our weekly research work. Blog address: http://summerprojectamity.blogspot.in/ and further studied the matlab image processing tool box Various Functions for our uitility  eg  bwareaopen studied to remove extra pixels etc.

References:

MATLAB DOCUMENTATION

Various websites And articles

Weekly Project Report (Week ONE)

Project Title :-Optical Character Recognition(OCR)

Team Members : 

Raghav Gogia & Karan Razdan
Btech(CSE),Pursuing  

Under Guidance of :

Ms. Shilpi Gupta

B.Tech(CSE), M.Tech(CSE Gold Medalist), MBA(HR & IT), PhD Pursuing
Assistant Professor - CSE department, Amity School of Engineering & Technology

Progress Made in Week One as on 11 May 2014:

For carrying out the process of optical character recognition (OCR), we needed to learn the details of Image processing. For that we need a language or a tool so we choose  MATLAB coz of following advantages offered by MATLAB

•  A very large (and growing) database of built-in algorithms for image processing and computer vision applications

• MATLAB allows you to test algorithms immediately without recompilation. You can type something at the command line or execute a section in the editor and immediately see the results, greatly facilitating algorithm development.

• The MATLAB Desktop environment, which allows you to work interactively with your data, helps you to keep track of files and variables, and simplifies common programming/debugging tasks

• The ability to read in a wide variety of both common and domain-specific image formats.

• A large user community with lots of free code and knowledge sharing

• The ability to auto-generate C code, using MATLAB Coder, for a large (and growing) subset of image processing and mathematical functions, which you could then use in other environments, such as embedded systems or as a component in other software.

We have been studying in detail, the various tools available in the MATLAB software package that would help us in the designing of OCR. 

We learned about the image processing tool box.  And studied the various function in it .

Some Important functions learnt

Image is loaded in memory using a imread command and processed using the imtool command (imshow can be used as alternative to display ). The imtool command is used to load a particular image and the various parameters can be used to display the image accordingly. It can be set to display a grayscale image within certain display range.

Example

I=imread(‘test.jpg’);

This command loads the image named ‘test’ in the current working directory and is referred by variable i

imtool (I)

The imtool opens the image viewer which has various functions. It can be used to import various images of different formats from the workspace. You can open using the file menu as well. The image viewer allows us to modify and work on the image in various manners.

We can look at the image in the form of a pixel grid using the ‘Inspect pixel values’ tool. This helps us look at the RGB values of each and every pixel individually in a very convenient manner. This picture is zoomed in so we can see each pixel value easily.

We can view information about the image such as the resolution and the colour format using the image information tool.

Contrast of the image can be adjusted too.

We can pan, move, zoom in and zoom out according to our convenience. We can measure distances between two points in terms of the number of pixels between those points.

The final tool is the crop tool. It is used to adjust the image according to the need. The unnecessary portion can be cropped off easily. The edited image can be saved in a format of our choice using this image viewer.