https://github.com/andrew2077/fall-detection
Time-series analysis for fall detection
https://github.com/andrew2077/fall-detection
Last synced: 3 months ago
JSON representation
Time-series analysis for fall detection
- Host: GitHub
- URL: https://github.com/andrew2077/fall-detection
- Owner: Andrew2077
- Created: 2023-08-01T15:11:23.000Z (almost 2 years ago)
- Default Branch: main
- Last Pushed: 2023-08-01T15:28:22.000Z (almost 2 years ago)
- Last Synced: 2025-01-20T09:46:29.860Z (5 months ago)
- Language: Jupyter Notebook
- Size: 3.38 MB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Fall Detection
Time-Series Analysis for Fall Detection
## Project Description
This project aims to detect falls from time-series data collected from a wearable device containing an accelerometer. The primary goal is to design a model capable of accurately detecting falls from the given time-series data.
## Data
The data was gathered from a single subject and exclusively tested on the subject. It comprises three different activities: walking, falling, and kneeling.
## Preprocessing
To ensure clean data, noise was manually removed. Labels were added to indicate the activity types correctly. Additionally, during the preprocessing phase, the derivative of the movement across all axes was considered for improved results.
## Model Design
The model design consists of a Bi-directional LSTM to analyze the movement sequence across all axes, followed by an MLP to classify the activity type based on the captured sequence.
## Results
Due to the limited data with only one subject, the model exhibited signs of overfitting to the subject.
## Quantization
Dynamic range quantization was employed to reduce the model size, and conversion to tflite format enabled the model's deployment on mobile devices.