Age prediction by Facial Features recognition using Yolo v4 Test phase, srs, design phase and source code final deliverable
Project Domain /
Category
Image Processing
Abstract /
Introduction
YOLO is an algorithm that uses neural networks to provide real-time object detection. This algorithm is popular because of its speed and accuracy. It has been used in various applications to detect traffic signals, people, parking meters, and animals. We will use yolo algorithm to predict age of persons based on different facial features.The first step will be to label the images, for this purpose we can use different available tools like LabelImg. Then we will setup environment and train our model and predict the age group.
Functional Requirements:
- First step is to select 500 images for each age group from the dataset total 7000 images, you can also select images from different datasets and generate a combined dataset. Following 14 age groups(classes) must be manually generated:
| 1-5 | 6-10 | 11-15 |
| 16-20 | 21-25 | 26-30 |
| 31-35 | 36-40 | 41-45 |
| 46-50 | 51-55 | 56-60 |
| 61-65 | 66-70 |
- You can write script to automatically copy the images to respective folder so the images can be easily annotated.
- Next step is to annotate the images according to the different age groups in yolo format using LabelImg.
- Once the dataset is annotated yolo environment will be set and model must be trained.
- You can set environment on your own machine or use google coolab.
- Model must be retrained if desired accuracy is not achieved by enhancing dataset or changing training parameters. Tools:
- Python
- jupyter notebook
- Yolo