Realtime-FallDetectionRNN / Git / Diff of /README.md

Models:

Robert-Orr/

Realtime-FallDetectionRNN

Downloads: 1

Diff of /README.md [857d1b] .. [ce2368]

Switch to unified view


# Real-time Fall Detection for RNN(AFD-RNN)

<p align="left">
<img src="https://github.com/chizhanyuefeng/Fall_Detection_for_RNN/blob/master/result/rnn.gif?raw=true", width="720">
</p>

result picture illustrate：

- The red,green,blue lines is acceleration sensor's x,y,z data。
- In the picture ,"correct" is the ground truth,"predict" is AFD-RNN network predict data
- Fall1、Fall2、Fall3 and Fall4 are represent Forward-lying,Front-knees-lying,Back-sitting-chair,Sideward-lying 

## AFD-RNN using RNN 
The sensors(acceleration and gyroscope sensor) is realtime to collect data,so we using rnn to detect the people movement.

## Requirenment
- TensorFlow >= 1.4
- python3
- matplotlib

## Class
Sitting,standing,stand to sit,sit to stand,upstairs,downstairs,lying,jumping,joging,walking and fall.

## Train and test

### 1.Train data
- The data collect frequence is 50Hz
- Need acceleration and gyroscope sensor

### 2.Before training
Put the train data to ./dataset/train/,and use kalman filter  to handle the data.


    python utils.py

### 3.Training
    
    python train_rnn.py
    
## 4.Testing
Put the test data to ./dataset/test/,and use kalman filter  to handle the data.


    python run_rnn.py
    
## Dataset

We using public dataset [MobileFall](http://www.bmi.teicrete.gr/index.php/research/mobiact) to train and test our net.

I upload the dataset at [Baidu网盘](https://pan.baidu.com/s/1arZMNPs1GzWrQf4beJFCSQ),if you cant download from [MobileFall](http://www.bmi.teicrete.gr/index.php/research/mobiact),you can try this

The final accuracy is 98.78%

	a/README.md		b/README.md
1	# Real-time Fall Detection for RNN(AFD-RNN)	1	# Real-time Fall Detection for RNN(AFD-RNN)
2		2
3	<p align="left">	3	<p align="left">
4	<img src="https://github.com/chizhanyuefeng/Fall_Detection_for_RNN/blob/master/result/rnn.gif", width="720">	4	<img src="https://github.com/chizhanyuefeng/Fall_Detection_for_RNN/blob/master/result/rnn.gif?raw=true", width="720">
5	</p>	5	</p>
6		6
7	result picture illustrate：	7	result picture illustrate：
8		8
9	- The red,green,blue lines is acceleration sensor's x,y,z data。	9	- The red,green,blue lines is acceleration sensor's x,y,z data。
10	- In the picture ,"correct" is the ground truth,"predict" is AFD-RNN network predict data	10	- In the picture ,"correct" is the ground truth,"predict" is AFD-RNN network predict data
11	- Fall1、Fall2、Fall3 and Fall4 are represent Forward-lying,Front-knees-lying,Back-sitting-chair,Sideward-lying	11	- Fall1、Fall2、Fall3 and Fall4 are represent Forward-lying,Front-knees-lying,Back-sitting-chair,Sideward-lying
12		12
13	## AFD-RNN using RNN	13	## AFD-RNN using RNN
14	The sensors(acceleration and gyroscope sensor) is realtime to collect data,so we using rnn to detect the people movement.	14	The sensors(acceleration and gyroscope sensor) is realtime to collect data,so we using rnn to detect the people movement.
15		15
16	## Requirenment	16	## Requirenment
17	- TensorFlow >= 1.4	17	- TensorFlow >= 1.4
18	- python3	18	- python3
19	- matplotlib	19	- matplotlib
20		20
21	## Class	21	## Class
22	Sitting,standing,stand to sit,sit to stand,upstairs,downstairs,lying,jumping,joging,walking and fall.	22	Sitting,standing,stand to sit,sit to stand,upstairs,downstairs,lying,jumping,joging,walking and fall.
23		23
24	## Train and test	24	## Train and test
25		25
26	### 1.Train data	26	### 1.Train data
27	- The data collect frequence is 50Hz	27	- The data collect frequence is 50Hz
28	- Need acceleration and gyroscope sensor	28	- Need acceleration and gyroscope sensor
29		29
30	### 2.Before training	30	### 2.Before training
31	Put the train data to ./dataset/train/,and use kalman filter to handle the data.	31	Put the train data to ./dataset/train/,and use kalman filter to handle the data.
32		32
33		33
34	python utils.py	34	python utils.py
35		35
36	### 3.Training	36	### 3.Training
37		37
38	python train_rnn.py	38	python train_rnn.py
39		39
40	## 4.Testing	40	## 4.Testing
41	Put the test data to ./dataset/test/,and use kalman filter to handle the data.	41	Put the test data to ./dataset/test/,and use kalman filter to handle the data.
42		42
43		43
44	python run_rnn.py	44	python run_rnn.py
45		45
46	## Dataset	46	## Dataset
47		47
48	We using public dataset [MobileFall](http://www.bmi.teicrete.gr/index.php/research/mobiact) to train and test our net.	48	We using public dataset [MobileFall](http://www.bmi.teicrete.gr/index.php/research/mobiact) to train and test our net.
49		49
50	I upload the dataset at [Baidu网盘](https://pan.baidu.com/s/1arZMNPs1GzWrQf4beJFCSQ),if you cant download from [MobileFall](http://www.bmi.teicrete.gr/index.php/research/mobiact),you can try this	50	I upload the dataset at [Baidu网盘](https://pan.baidu.com/s/1arZMNPs1GzWrQf4beJFCSQ),if you cant download from [MobileFall](http://www.bmi.teicrete.gr/index.php/research/mobiact),you can try this
51		51
52	The final accuracy is 98.78%	52	The final accuracy is 98.78%