Sign Language Detection using YOLOv5

Introduction

The Sign Language Detection project focuses on the recognition and interpretation of hand gestures within sign language. This comprehensive solution utilizes the YOLOv5 object detection model to effectively identify common sign language expressions like "Hello," "I love you," "Yes," "No," and "Please." By doing so, it aims to enhance communication and foster a deeper understanding between individuals who use sign language and those who may not be familiar with it.

Model Architecture

The project utilizes the YOLOv5 object detection model, a popular and efficient deep learning model for real-time object detection. Visit YOLOv5 Github Repo

Dataset Creation

To make YOLOv5 work, we need to teach it what the sign gestures look like. We do this by creating our own dataset and also integrating it with the publicly available datasets. For creating our own dataset, ran this code in Python. ``` import os import cv2 import time import uuid

IMAGEPATH='CollectedImages' labels=['Hello','Yes','No','Thanks'] numberofimages=4 for label in labels: imgpath = os.path.join(IMAGEPATH, label) os.makedirs(imgpath) cap=cv2.VideoCapture(0) print('Collecting images for {}'.format(label)) time.sleep(5) for imgnum in range(numberofimages): ret,frame=cap.read() imagename=os.path.join(IMAGE_PATH,label,label+'.'+'{}.jpg'.format(str(uuid.uuid1()))) cv2.imwrite(imagename,frame) cv2.imshow('frame',frame) time.sleep(2)

    if cv2.waitKey(1) & 0xFF==ord('q'):
        break
cap.release()

```

Formatting Data

After Collecting Data we have to format this image Yolo format using online tools Visit online tools link for formatting image

train | images | labels |valid | images | labels

Training Data

python3 train.py --img-size 640 --batch 16 --epochs 100 --data custom.yaml --weights yolov5s.pt --nosave --cache Optimizer stripped from runs/train/exp/weights/last.pt,

python3 detect.py --weight runs/train/exp/weights/last.pt --img 640 --conf 0.25 --source 0

Usage

To use this Sign Language Detection project, follow these steps:

REFERENCES

• Annotation Tool Blog
• Medium Blog
• Roboflow Blog

https://medium.com/@mycodingmantras/building-a-real-time-object-detection-and-tracking-app-with-yolov8-and-streamlit-part-2-d1a273592e7e

https://github.com/CodingMantras/yolov8-streamlit-detection-tracking/tree/master/videos

https://github.com/CodingMantras/yolov8-streamlit-detection-tracking

https://lalodatos.medium.com/building-your-own-real-time-object-detection-app-roboflow-yolov8-and-streamlit-part-4-16a025c7240c

https://github.com/xugaoxiang/yolov5-streamlit/blob/main/requirements.txt http://localhost:8501/ python3 train.py --img-size 416 --batch-size 16 --epochs 50 --data custom.yaml --cfg models/yolov5s.yaml --weights yolov5s.pt

pip3 install streamlit
pip3 install pytube
pip3 install lap install pytube install --no-cache "lapx>=0.5.2"