Image Segmentation and Dimension Analysis

Introduction

Automated dimension analysis from an image is of immense importance in today's world. Current projects present in this domain do not provide a robust and easy to use UI. In the following project creates a robust software that works for most of the pictures while keeping in mind the ease of access. At the present stage this provides all the basic dimension that are important.

The code was tested on Matlab 2017a.

• Please refer to Report.pdf for detailed analysis.
• Please refer to lab.pdf for about the project.
• Please refer to results for segmentation results using various methods.

Directory Structure

---code

|
|---RunAll.m
|---Find_Bound_Box.m
|---Calculate_Boundary.m

---results

|
|---Active
|---Inbuilt
|---Lazy
|---Morphological

---lab.pdf

---README.md

---Report.pdf

---Literature_Review.pdf

To Run

RunAll( RGB, method) - Generates the boundary box around and calculate the bounding box dimension and boundary dimension Specify method as 'lazy'(default) , 'active contour', 'morphological', 'matlab'

e.g.

RunAll(imread('data/Phone 1.jpg'),'lazy')

RunAll(imread('data/Phone 2.jpg'),'active contour')

RunAll(imread('data/shoe.jpg'),'morphological')

Developed By

Naman Goyal (2015CSB1021) Koustav Das (2015CSB1017)