# keras-Retinanet-Safe-Hat-Detection

**Repository Path**: david12/keras-Retinanet-Safe-Hat-Detection

## Basic Information

- **Project Name**: keras-Retinanet-Safe-Hat-Detection
- **Description**: No description available
- **Primary Language**: Unknown
- **License**: Not specified
- **Default Branch**: master
- **Homepage**: None
- **GVP Project**: No

## Statistics

- **Stars**: 0
- **Forks**: 0
- **Created**: 2020-04-01
- **Last Updated**: 2020-12-19

## Categories & Tags

**Categories**: Uncategorized

**Tags**: None

## README


keras-retinanet-Hat-Detection
======
A safe hat detection model using keras-retinanet

****
	
|Author|LCorleone|
|---|---
|E-mail|lcorleone@foxmail.com


****

## Requirements
please see the original github [keras-retinanet](https://github.com/fizyr/keras-retinanet)  

## Notes
* Sorry I can not provide the training dataset, so only the trained model is available.
* The safe hat seems to be relatively small, by now I didn't figure out how to modify the anchor settings correctly in keras-retinanet. Indeed I try but the performance is not satisfactory. If anyone know about this, feel free to contact me!
* The trained model is available here [BAIDU PAN](https://pan.baidu.com/s/1EkiJhNivu9Pjdr350kM6UA)

## Results
![image](https://github.com/LCorleone/keras-retinanet-Hat-Detection/blob/master/results/video1-20180724_VID_20180724_120103_360.jpg)