https://github.com/faisalahmedbijoy/image_cryptography_with_autoencoders

Deep learning in image Cryptography
https://github.com/faisalahmedbijoy/image_cryptography_with_autoencoders

cryptography deep-learning image image-processing security

Last synced: about 2 months ago
JSON representation

Deep learning in image Cryptography

• Host: GitHub
• URL: https://github.com/faisalahmedbijoy/image_cryptography_with_autoencoders
• Owner: FaisalAhmedBijoy
• Created: 2023-08-19T03:42:20.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-06-26T19:00:22.000Z (about 1 year ago)
• Last Synced: 2025-02-27T11:46:45.837Z (4 months ago)
• Topics: cryptography, deep-learning, image, image-processing, security
• Language: Python
• Homepage:
• Size: 31.6 MB
• Stars: 3
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# Image_Cryptography_with_Autoencoders

Deep learning in image Cryptography

Image encryption and decryption using a chaotic map sequence and an autoencoder. The repository includes code for generating chaotic map sequences, shuffling and deshuffling images, preparing datasets, and computing performance metrics.

## Environment setup

```bash

conda env create --file environment.yml

```

**Install packages**

```bash

keygen==0.0.3

matplotlib==3.3.3

numpy==1.23.1

numpy==1.23.2

opencv_python==4.4.0.46

scikit_learn==0.23.2

skimage==0.0

tensorflow==2.3.0

```

To install the required packages, run the following command:

```bash

pip install -r requirements.txt

```

## 1. Chaotic map sequence generation

To generate the chaotic map sequence, run the following command:

```bash

python generate_chaotic_map_sequence.py

```

## 2. Shuffle and Deshuffle image

To shuffle and deshuffle an image, run the following command:

```bash 

python shuffling_deshuffling_images.py

```

**Input image**



**Shuffle image** 

![](images/shuffled_deshuffled_image/faisal_shuffled_image.png)

**Deshuffle image**

![](images/shuffled_deshuffled_image/faisal_deshuffled_image.png)

## 3. Encryption and Decryption of image using chaotic map sequence

To encrypt and decrypt an image using chaotic map sequence, run the following command:

```bash

python image_encryption_decryption.py

```

**Encryption of image**

![](images/encrypted_decrypted_images/faisal_encrypted_image.png)

**Decryption of image**

![](images/encrypted_decrypted_images/faisal_decrypted_image.png)

## 4. Dataset preparation 

To prepare the dataset, run the following command:

```bash

python data_loader.py

```

## 5. Autoencoder for image encryption and decryption

To use the autoencoder for image encryption and decryption, run the following command:

```bash

python auto_encoder.py

```

**Model architecture**

![](images/model_architecture_and_performances/autoencoder_architecture.png)

**Model training**

```bash

python train.py

```

**Model loss performance graph**

![](images/model_architecture_and_performances/loss_graph.png)

## 6. Inferencing the model

*original_vs_compressed_vs_reconstruction*

```bash

python inference.py

```

**Original image**

![](images/model_architecture_and_performances/original_image.png)

**encoder compressed image**

![](images/model_architecture_and_performances/compressed_encoded_image.png)

**Decoder decompressed image**

![](images/model_architecture_and_performances/decompressed_decoded_image.png)

## 7. Performance metrics

- Structural Similarity Index (SSIM)

```bash

python structural_similarity_SSIM_calculation.py

```

- Number of pixel change rate (NPCR)

```bash

python nnumber_of_pixel_change_rate_NPCR_comparision.py

```

- NPCR computation

**image 1 output**

![](images/NPCR_images/NPCR_difference_1_faisal.png)

**image 2 output**

![](images/NPCR_images/NPCR_difference_2_faisal.png)

- Unified Average Changing Intensity (UACI)

```bash

python unified_average_changing_intensity_UACI_comparision.py

```

![](images/UACI_images/UACI_difference_faisal.png)

## 8. Salt and paper noise

    

 ```bash

   python salt_and_pepper_noise.py

```

**Noisy image**

![](images/noisy_images/faisal_noisy.png)