Shape Detection using Image Processing Tools

Shape Detection using Image Processing Tools

The project, “shape detection using Image processing tools”. Matlab was the chosen software that recognises the shapes of 2D figures loaded unto it and names them appropriately. The shape detection is achieved by the software detecting the edge detection method such as the canny method and finding the area, perimeter of the shape in order to detect the shapes

Week 5

Summary of week's activities

• Poster design and printing
• Report writing
• Finalizing the code and testing

Final Result 

Conclusion 

The shapes detection project cut across five weeks of both team and individual work, the project was entirely new to every member of the group, but good team work and individual commitment made the project a success.

The first week of the project was when it all began as we didn't have even an unofficial meeting before the first week of the project.

The project deliverable was divided efficiently among the three members; code design, GUI design and project manager/technical writer which were the the responsibilities of Ashwin Varughese, Mohammad Houtari and Telema Apaemi respectively. Nonetheless the different allocations,  we were all free to share ideas on how each subdivision of the project could be done better, owing to this we decided to open a group chat to facilitate ideas sharing and keep each other updated on in time progress and challenges.

                                                                        Limitations faced by the project 

• Only a finite set of geometric shapes can be detected
• The images have to be hand made using  Paint drawing tools or any software
• The software won't detect if the area is too big or too small.

  

week 4

Summary Of Weeks activities

• Report writing 
• Poster blueprint 
• GUI design and in-cooperating it with the software 
• Further testing and final code editting

week 3

Week 3

Summary of week's activities:

This week we were able to start our report and extend our set of shapes to include ellipse and hexagon
Testing the code, we were able to observe two problems;

methods extended to detect ellipse and hexagon

• images with rough edges couldn't be detected properly
• the program could neither detect very small images nor very large one

The former was solved with the use of a matlab built in function bwareaopen();
The link below shows the syntax and use for the bwareaopen() function.
http://uk.mathworks.com/help/images/ref/bwareaopen.html

The later couldn't be solved within the time slot and we resolved on doing more research on how to tackle the problem 

week 2

Summary of Weeks activities:

From the resolution on the previous week's meeting, useful research was done and the first phase of coding was started.

At the end of the week's meeting we were able to use matlab's image processing tools to detect shapes ; circles, rectangle, square, triangle and diamond.

The week's challenge was to extend this to a wider range of shapes and a matlab method to reduce noise from images .

diagram showing shapes identified by the software

week 1

On today's lab session the group did the following;

• familiarized ourselves with the project deliverable
• familiarized ourselves with matlab interface and algorithms
• researched on previous years work on the subject and other relevant sources
• distribution of task among team 

The team resolved on doing more individual research on the project with the aim of getting the best and most feasible approach to handling the project.
The team also decided on the task for the following week following the progress of the current week which was to start coding and testing at the same time.