đ Bachelor thesis
Forensic Tool for Detecting Child Pornography Using Deep Neural Networks
The tool, developed as part of a bachelor's thesis after the Faculty of Applied Computer Science, is used to detect child pornography.
The detection methods are based on convolutional neural networks.
The detection methods are based on convolutional neural networks.
Part of the application | Technologies used |
---|---|
Neural model | TensorFlow, Keras |
Model inference | ONNX Runtime |
GUI application | Java, Swing |
Web presentation | Chakra UI, Next.js, Streamlit |
Input image | 224x224 |
Layers | 316 |
Parameters | 22M |
Accuracy | 87.068% |
Main repository with Jupyter notebooks and toolkits
filipsedivy/bachelor-rajce-crawler
Crawler to create API for rajce.idnes.cz
If you have any questions, you can contact us at đ§ [email protected].