Object Detection java

Overview
How to Install and Run the Project
File Contents
Introduction
- Neighbourhood
- Connectivity
Input images of objects
Program Screenshots

Overview

It is a Java based GUI desktop application which “detects the objects in a binary image using the concept of Neighborhood and connectivity”. More specifically it implements the concept of "M-Connectivity" of Digital Image Processing.

How to Install and Run the Project

Method # 1: Download 'Executable (exe)' file from 'PORTABLE APPLICATION' Folder (Easy & Portable)

            No need to install anything but it is for WINDOWS OS

Method # 2: Download and Load Project in Eclipse IDE to Run it

            Eclipse IDE & JVM will be required.

File Contents

src/Application/ObjectDetection.java includes the implementation of neighbourhood & connectivity concept (main logic).
src/com/test/jfilepicker package includes source files for GUI components.
src/com/test/jfilepicker package/images package includes the binary images.

Introduction

1-Neighborhood

A Digital Image comprises of a 2D matrix, and every element is called as Pixel. The Image is denoted by f(x,y) and any pixel can be denoted by small letter, say p, q … Every Pixel has some Neighbors located around it.

Different types of neighbors include: 4 Neighbors D Neighbors 8 Neighbors

4-Neighbors

A pixel ’p’ located at coordinate value (x,y) has 4 neighbors, which are adjacent to ‘p’. The coordinate values can be determined as: (x+1,y) (x-1,y) (x,y+1) (x,y-1) by N4(p) 4 Neighbors are denoted

4-Neighbors

(ii) D-Neighbors

A pixel ’p’ located at coordinate value (x,y) has D neighbors, which are located in Diagonal of ‘p’. The coordinate values can be determined as: (x+1,y+1) (x+1,y-1) (x-1,y+1) (x-1,y-1) 8 Neighbors are denoted by N8(p)

D-Neighbors

(ii) 8-Neighbors

A pixel ’p’ located at coordinate value (x,y) has 8 neighbors, which are combination of 4 and D Neighbors. 8 Neighbors are denoted by N8(p)

(x+1,y) (x-1,y) (x,y+1) (x,y-1) (x+1,y+1) (x+1,y-1) (x-1,y+1) (x-1,y-1)

8-Neighbors

Connectivity

Connectivity between pixels is an important concept used in establishing boundaries of objects and components of regions in an image.

As we can identify different objects in Image depending upon the Intensity values of Pixels.

So all Pixels comprising an object would have Same or Close values.

To establish whether two pixels are connected, it must be determined

If they are adjacent in some sense (say, if they are 4-neighbors) . and If their grey levels satisfy a specified criterion of similarity.

The specified Criteria (Intensity Range ) can be defined by defining a set ‘V’ at start. For V={2} Only those Pixel can be connected which have value =2 For V={1,2} Only those Pixel can be connected which have value =1 or 2. The specified Criteria (Intensity Range ) can be defined by defining a set ‘V’ at start.