Image Search Engine

Image Search Engine:

In normal Search Engine like Google, DuckDuckGo, Bing(lol) we use text to search the database(or Web), we simply enter a few keywords related to the content that we want to find and BAM the result. But in Image Search Engine we use image instead of text to query the Database(or Web).

Types Of ISE (Image Search Engine):

In general, there tend to be three types of image search engines:
(1) search by meta-data (2) search by example (3) hybrid approach of the two

Search By Meta-Data

Search by meta-data systems rarely examine the contents of the image itself. Instead, they rely on textual clues such as
(1) Manual annotations and tagging performed by humans (2) Automated contextual hints, such as the text that appears near the image on a webpage

Search By Example

Search by example systems, on the other hand, rely solely on the contents of the image — no keywords are assumed to be provided. The image is analyzed, quantified, and stored so that similar images are returned by the system during a search.

Hybrid Approach Of The Two

hybrid image search engine includes both text keywords along with features extracted from the images.

Dataset Used To Test The Program

ftp://ftp.inrialpes.fr/pub/lear/douze/data/jpg1.tar.gz
ftp://ftp.inrialpes.fr/pub/lear/douze/data/jpg2.tar.gz

The Task

The task for this program is to be used as a Image Search Engine for a Database/Dataset

How To Use

I recomend you read the code (Code is well Defined and Commented) To understand How this program works (Or Wait for my Youtube tutorial/video when this project is compeleted)

Step 1:

install Python3, pip and git

Linux

# For Debian and Ubuntu Users
sudo apt install python3
sudo apt install python3-pip
sudo apt install git

# For Arch Users
sudo pacman -S python # Default python in arch is python3 not python2.7
sudo pacman -S python-pip
sudo pacman -S git

# For Fedora Users
sudo dnf install python3
sudo dnf install python3-pip
sudo dnf install git

Compile It From Source : https://www.unixmen.com/howto-install-python-3-x-in-ubuntu-debian-fedora-centos/

MacOS

From This link Download the latest python3 for MacOS: https://www.python.org/

Or Use Homebrew

# To install Homebrew on MacOS
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/master/install.sh)"

# To install Python3 on MacOS using Homebrew
brew install python
# or brew install python3
brew install git

this will autometically install pip as well you might see there is no pip then try to use pip3

Windows

Got to this link and Download and install python3 for windows : https://www.python.org/ (I personally preffer linux-subsystem)

Step 2

Install python requirements using pip/pip3

pip install -r requirements
# you can also use sudo in mac and linux to install as root but i don't suggest you do it

Step 3

Download the git-repo

git clone https://gitlab.com/saurabhtanwar/image-search-engine.git
# then go into git directory
cd image-search-engine

Step 4

Index your Image Dataset for learning

python index.py --dataset <image-dataset-directory> --index index.csv # index.csv is out file for indexing

Step 5

Now Use Image Search Engine

python search.py --index index.csv --query <Loacation of image to use a query/searching perameter> --result-path <image-dataset-directory>

This project is not fully compeleted

(As per my vision… It works but i want to make it so that it can search several shoping wesites and then arrenge data or products using 2 perameters i.e. More accurate products and more cheaper products)