image-transformations-app

A streamlit app to visualize images/video and apply basic image transformations to get some insights. The app is built on streamlit and on the backend, opencv functions are called to be applied on the image.

Usage

A video could be analysed with the following command

streamlit run app.py --data path_to_video_file

or a folder of images could also be viewed

streamlit run app.py --data path_to_folder_containing_images

Requirements

There are a few requirements before we can fully start using the list

opencv
streamlit
numpy 
glob
matplotlib
argparse

Features

The repo has the following features to play with

Basic Functions

This functions include some of the commonly used ones

1. Conversion to Grayscale
2. Conversion to HSV
3. Displaying Histogram
4. Histogram equalization
5. Rotation of the image

Thresholding functions

It also provides simple thresholding functions that are part of the opencv. These can be applied to a specific channel and this can be chosen by the options.

Applying various filters

There is a provision to apply some of the common filters like gaussian kernels or a median filter. Even kernel size could be adjusted.

Applying Color transformations

Some of the common color transformations like contrast and brightness could be tweaked in the app.

Applying Image transformations

Apart from basic thresholding, even some morphological transformations can also be visualized.