The open-source Hard Hat Detection model yolos-tiny - A must for object detection in workplace safety monitoring

Yolos Tiny Hard Hat Detection

Developed by DunnBC22

A fine-tuned hard hat detection model based on YOLOS-tiny, designed for object detection tasks in workplace safety monitoring scenarios.

Object Detection

Transformers

EnglishOpen Source License:Apache-2.0 #Hard Hat Detection #Workplace Safety Monitoring #Small Object Detection

Downloads 13

Release Time : 8/4/2023

Model Overview

This model is a fine-tuned version of hustvl/yolos-tiny on a hard hat detection dataset, primarily used to detect whether personnel in images or videos are wearing safety helmets. It is suitable for construction sites, factories, and other locations requiring safety protection.

Model Features

Efficient Detection

Optimized based on the YOLOS-tiny architecture, achieving fast inference while maintaining high detection accuracy.

Workplace Safety Application

Specifically optimized for hard hat detection scenarios, suitable for safety monitoring needs in construction sites and factories.

Multi-scale Detection

Capable of detecting targets of various sizes, including small, medium, and large hard hat wearing scenarios.

Model Capabilities

Image Object Detection

Hard Hat Recognition

Real-time Monitoring Analysis

Use Cases

Workplace Safety Monitoring

Construction Site Safety Monitoring

Automatically detects whether workers on-site are wearing hard hats to prevent safety accidents.

Achieved 74.7% AP@0.5 accuracy on the test set.

Factory Safety Inspection

Identifies violations of not wearing hard hats through video surveillance.

Achieved 52.1% AP@0.50:0.95 for large object detection.

🚀 yolos-tiny-Hard_Hat_Detection

This model is a fine - tuned version of hustvl/yolos-tiny on the hard-hat-detection dataset, which can be used for object detection in workplace safety scenarios.

🚀 Quick Start

This model is a fine - tuned version of hustvl/yolos-tiny on the hard-hat-detection dataset.

📚 Documentation

Model description

For more information on how it was created, check out the following link: https://github.com/DunnBC22/Vision_Audio_and_Multimodal_Projects/blob/main/Computer%20Vision/Object%20Detection/Hard%20Hat%20Detection/Hard_Hat_Object_Detection_YOLOS.ipynb

Intended uses & limitations

This model is intended to demonstrate my ability to solve a complex problem using technology.

Training and evaluation data

Dataset Source: https://huggingface.co/datasets/keremberke/hard-hat-detection

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 5e-05
train_batch_size: 8
eval_batch_size: 8
seed: 42
optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
lr_scheduler_type: linear
num_epochs: 8

Training results

Metric Name	IoU	Area	maxDets	Metric Value
Average Precision (AP)	IoU=0.50:0.95	all	maxDets=100	0.346
Average Precision (AP)	IoU=0.50	all	maxDets=100	0.747
Average Precision (AP)	IoU=0.75	all	maxDets=100	0.275
Average Precision (AP)	IoU=0.50:0.95	small	maxDets=100	0.128
Average Precision (AP)	IoU=0.50:0.95	medium	maxDets=100	0.343
Average Precision (AP)	IoU=0.50:0.95	large	maxDets=100	0.521
Average Recall (AR)	IoU=0.50:0.95	all	maxDets=1	0.188
Average Recall (AR)	IoU=0.50:0.95	all	maxDets=10	0.484
Average Recall (AR)	IoU=0.50:0.95	all	maxDets=100	0.558
Average Recall (AR)	IoU=0.50:0.95	small	maxDets=100	0.320
Average Recall (AR)	IoU=0.50:0.95	medium	maxDets=100	0.538
Average Recall (AR)	IoU=0.50:0.95	large	maxDets=100	0.743

Framework versions

Transformers 4.31.0
Pytorch 2.0.1+cu118
Datasets 2.14.3
Tokenizers 0.13.3

📄 License

This model is licensed under the Apache 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご