Tata Innoverse - Solverhunter Image Stitching software built Upon python using OpenCV
Test Dataset : https://www.dropbox.com/sh/qa9gwcpp1etgkd7/AAASYeaUxij4a2kTiOd0NSVna?dl=0 Results : https://drive.google.com/open?id=14KnZ6C9RpAOMN2BmqXRex8UJ7mWVfP4g
Requirements
- Python >= 3.5
- pip
Set Up
- Firstly download the project zip file and extract its contents.
- Open Command Prompt (with administrative privileges) and navigate to the project folder.
- Run the following in the Command Prompt (or Terminal on Linux) to install all the required dependencies pip install -r requirements.txt
Run the Project
- Open the command Prompt (or Terminal on Linux) in the Project folder.
- Place the test dataset images in datasets/images folder
- Run the following in the Command Prompt (or Terminal) python ImageMosaic.py
- Final Image is saved as finalResult.png inside results folder.
Minimum Specifications
- 4GB RAM
- At least 2GB Free Disk Space For storing temporary files
Our Test Bench
- 4GB RAM with Core i5 3.6 ghz (Desktop) ~ 10 mins
- 8GB RAM with Core i5 2.6 ghz (Laptop) ~ 20 mins
Note
- The software is not compatible with Python 2.
- All the images must have EXIF data and XMP data with them. (Mostly drone captured images already have these as metadata.)
- If the software fails to import OpenCV, then please run the project over a virtual environment. Please follow up this link to setup virtualenv.