Computer Vision

Overview

• This repository contains image operation implementations. For more detail please check the Features section.

Installation

• You can clone this repo with following command.

$ git clone https://github.com/Kadiryanik/ComputerVision

Compile & Run

$ cd ComputerVision/
$ make # for compile
$ ./test -h # for printing usage

Features

• Converting rgb image to grayscale form
• Converting rgb/grayscale image to binary form
• Plot histogram (with python) while converting to binary form
• Drawing shapes into image
• Cropping image
• Applying mask into image
• Applying morphologhy into image (dilation|erosion|open|close)
• Identifying regions on image

Drawing

• Drawing some shapes.

$ ./test -i images/backup/rgb-lena.bmp -d draws/face.txt

Input	Output

Morphology

• Applying morphology open.

$ ./test -i images/backup/numbers/number-w-noise-01.bmp -M open

Input	Output

---

Identifying Regions

• Calculating (learning) the features.

$ ./test -f learn -i db/numbers-class-image-db.txt -o db/my-numbers-features-db.txt

• Testing mixed photo with calculated features.

$ ./test -f test -i db/my-numbers-features-db.txt -T images/backup/numbers/number-w-noise-01.bmp

• Please download output image to view in high resolution.

Input	Output

TODO

• Add zoom in/out support
• Add jpeg support
• Add png support
• Add real photos for test identification feature
• Add user defined colors support with formatted input files
• Add Color-Class relation into image corner
• Add match percentage into region corner

Maintainer

• Kadir Yanık - kdrynkk@gmail.com

Free Software, Hell Yeah!