https://github.com/dongju93/file-to-image-for-cnn

Convert files to binary, then transform them into RGB images. Subsequently, use these RGB images to train CNN models in a supervised manner for malware detection.
https://github.com/dongju93/file-to-image-for-cnn

convolutional-neural-networks deep-learning javascript jupyter-notebook machine-learning python pytorch supervised-learning tensorflow typescript

Last synced: 10 days ago
JSON representation

Convert files to binary, then transform them into RGB images. Subsequently, use these RGB images to train CNN models in a supervised manner for malware detection.

• Host: GitHub
• URL: https://github.com/dongju93/file-to-image-for-cnn
• Owner: dongju93
• Created: 2023-08-30T22:46:03.000Z (about 1 year ago)
• Default Branch: main
• Last Pushed: 2023-09-29T10:00:44.000Z (about 1 year ago)
• Last Synced: 2024-01-26T09:19:44.451Z (10 months ago)
• Topics: convolutional-neural-networks, deep-learning, javascript, jupyter-notebook, machine-learning, python, pytorch, supervised-learning, tensorflow, typescript
• Language: Jupyter Notebook
• Homepage:
• Size: 21.5 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# GOAL

1️⃣ Convert the malicious file to a .PNG image - ✅  

2️⃣ Image size is 256x256, RGB color - ✅  

3️⃣ Create deep learning CNN model - ✅  

4️⃣ Feed the model with an image of the suspicious file (steps 1-2) - ✅  

5️⃣ Derive how similar the result is to the trained (malicious) content in %. - ⬜️  

6️⃣ Determine malicious if it is above a certain similarity %, and not malicious if it is below. - ⬜️  

7️⃣ (option) Visualize the model training and results - ⬜️

< based on this [article](https://ieeexplore.ieee.org/document/8887303) >

Change file into images

```

// typescript compile

// tsc file_into_images.ts

// run javascript

node file_into_images.js

```

[Pytorch](https://pytorch.org)

```

// for run this code

// open pytorch_cnn.ipynb and select python version than click Run All

// dependencys are required

pip install torch torchvision Pillow matplotlib

```

[Tensorflow](https://www.tensorflow.org/)

```

// for run this code excute command below

node tensorflow_cnn.js

// dependencys are required

npm i @tensorflow/tfjs-node pngjs

```