Title 1: LSTM-Based Human Fall Detection using Thermal Array Sensor (done)
Title 2: Real-Time Fall Detection System using Embedded Machine Learning
[1] Setting up test bed for thermal array dataset generation
[2] Define scenario of the dataset generation using thermal array sensor. The scenario are:
- Capture 2 kinds of movement: falling from walking and falling from sitting
- each movement monitored in a) room without AC and b) room with AC turned on
- each monitoring is using three kinds of sensor resolutions once at a time (4x4 sensor, 4x4 H sensor, and 32x32 sensor)
- monitor each movement twice
[3] Collect the dataset using the defined scenario
[4] Preprocess the obtained dataset into csv and xlsx files from txt and label the dataset.
[1] Train various ML method using the obtained dataset to the performance of ML method and which one is the best thermal array resolution from the obtained dataset
[2] Paper survey on embedded machine learning using Arduino
IoT applications for Smart Home
[1] Setting up test bed for thermal array dataset generation
[2] Define scenario of the dataset generation using thermal array sensor. The scenario are:
- Capture 2 kinds of movement: falling from walking and falling from sitting
- each movement monitored in a) room without AC and b) room with AC turned on
- each monitoring is using three kinds of sensor resolutions once at a time (4x4 sensor, 4x4 H sensor, and 32x32 sensor)
- monitor each movement twice
[3] Collect the dataset using the defined scenario
[4] Preprocess the obtained dataset into csv and xlsx files from txt and label the dataset.
[1] Evaluate which thermal array resolution suits the fall detection purpose
[2] Train various ML methods to investigate the best performance of the given dataset.
Finish dataset generation using thermal array sensor and finish introduction part of summer intensive journal
Submit conference and/or iot journal page