https://github.com/joeyism/py-image-comparer

Compares two images using Siamese Network (machine learning) trained from a Pytorch Implementation
https://github.com/joeyism/py-image-comparer

compare images network pytorch siamese siamese-network torch

Last synced: 3 months ago
JSON representation

Compares two images using Siamese Network (machine learning) trained from a Pytorch Implementation

• Host: GitHub
• URL: https://github.com/joeyism/py-image-comparer
• Owner: joeyism
• License: apache-2.0
• Created: 2021-07-10T20:09:43.000Z (over 3 years ago)
• Default Branch: master
• Last Pushed: 2021-07-27T13:48:17.000Z (over 3 years ago)
• Last Synced: 2024-08-09T22:35:58.545Z (6 months ago)
• Topics: compare, images, network, pytorch, siamese, siamese-network, torch
• Language: Python
• Homepage:
• Size: 197 KB
• Stars: 10
• Watchers: 4
• Forks: 0
• Open Issues: 1
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# py-image-comparer

Compares two images using [Siamese Network](https://www.cs.cmu.edu/~rsalakhu/papers/oneshot1.pdf) (Machine Learning) trained from a [Pytorch Implementation](https://github.com/joeyism/siamese-pytorch)

## Installation

To install, run

```bash

pip install image-comparer

```

## Usage

### CLI

```bash

image-compare

```

which wil show the follow help screen

```

usage: image-compare [-h] [--threshold THRESHOLD] Image1-Path Image2-Path

```

For example, you can compare two images with

```bash

image-compare tests/images/kobe.jpg tests/images/kobe2.jpg 

```

which gives the result

```

kobe.jpg and kobe2.jpg are not similar

```

### Programmatically

With PIL

```python

import image_comparer

from PIL import Image

image = Image.open("test/kobe.jpg")

image2 = Image.open("test/kobe2.jpg")

image_comparer.is_similar(image, image2, threshold=0.5)

```

or with OpenCV

```python

import image_comparer

import cv2

image = cv2.imread("test/kobe.jpg")

image2 = cv2.imread("test/kobe2.jpg")

image_comparer.is_similar(image, image2, threshold=0.5)

```

## API

### Methods

`is_similar(image1: Union[Image.Image, np.ndarray], image2: Union[Image.Image, np.ndarray], threshold=0.5)`: Checks if the two images are similar based on the reshold passed

`calculate_score(image1: Union[Image.Image, np.ndarray], image2: Union[Image.Image, np.ndarray])`: Calculates the score between the two images. The higher the score, the more closely the two images are related.

## Development

### Installation

```bash

pip install -r requirements-test.txt

```

### Tests

To run tests, run

```bash

pytest

```