|
|
--- |
|
|
tags: |
|
|
- image-classification |
|
|
- transfer-learning |
|
|
- keras |
|
|
- tensorflow |
|
|
- security |
|
|
library_name: tensorflow |
|
|
--- |
|
|
|
|
|
# Fine-Tuned Weapon vs. Not-Weapon Classifier (VGG16) |
|
|
|
|
|
This repository contains a fine-tuned image classification model designed to distinguish between **"Weapon"** and **"Not-Weapon"** classes in images. The model was built using Transfer Learning on the VGG16 architecture. |
|
|
|
|
|
## Model Structure and Training |
|
|
|
|
|
* **Base Architecture:** VGG16 (pre-trained on ImageNet). |
|
|
* **Training Method:** Transfer Learning (Feature Extraction) followed by **Fine-Tuning**. |
|
|
* **File:** `weapon_classifier_final_tuned.keras` (This is the final, fine-tuned model file). |
|
|
* **Input Size:** Images must be resized to **(224, 224)** pixels before prediction. |
|
|
* **Output Activation:** Sigmoid (yielding a single probability value between 0 and 1). |
|
|
|
|
|
## How to Use the Model (Inference) |
|
|
|
|
|
To use this model, you need a Python environment with TensorFlow installed. |
|
|
|
|
|
### 1. Installation |
|
|
|
|
|
First, ensure you have the required libraries: |
|
|
|
|
|
```bash |
|
|
pip install tensorflow numpy Pillow |
