in Figure 1, the crib uses a set of sensors to measure the baby’s
weight, bed wetting, and, most importantly, analyses the infant’s
sentiment. While most sentiment analysis methods rely on
natural language processing,
text analysis, computational
linguistics, or bio statistics to systematically identify, extract,
quantify, and study emotional states and subjective information
[6],
we suggest to analyse babies’ sentiment by analysing their
crying patterns. To the best of our knowledge, this is the first
work that aims to understand infants’ sentiment based on their
crying. Moreover, our smart grip is able to trigger various
actions that parents would perform such as playing comforting
sound, or shaking the crib, and
additionally can inform the
parents about their infant’s needs.
The paper is structured as follows. In Section II, we provide
an overview of the system architecture. Section III outlines the
signal processing algorithm to analyze crying patterns. In
Section IV, we summarize the implementation of our system.
The experiment is introduced in Section V. Section VI
concludes this work.
II.
S
YSTEM
D
ESIGN
A.
System Architecture
As shown in Fig. 2, the system is divided into three
components: hardware (mainly Raspberry Pi and sensors),
server and mobile application.
•
Hardware: The hardware components collects all data
for processing. The main parts are various sensors to
record different signals and a Raspberry Pi [7] which is
used for processing. The Raspberry Pi has a small and
powerful microcontroller that
can handle complex data
processing tasks. In addition, the Raspberry Pi also has
many native sensor accessories that can be used for data
processing tasks. Its built-in wireless network module
enables it to interact with the server in real time.
•
Server: The server serves as backend of the system
architecture as it is the central hub used for data exchange.
It contains various servlets that are employed to record
infants’ details such as weight and to present this
information to the parents. The data is stored in a
MySQL database [8].
•
Mobile Application: The mobile Application allows
parents to display the data that is stored on the server.
Sensing
Crying
Bed-wetting
Weight
Logging
Sentiment analysis
Alert
Playing
music
Shaking
Alert
Alert
Alert
comfort
Hungry
Pain
Sleepy
Receive
alert
message
Take
actions
feeding
comfort
lull babies
to sleep
change
disapers
Figure 1. Functions of Smart Crib
The app runs in the background and frquently requests
data from theserver. A voice
command is triggered to
draw parents’ attention if needed.
B.
System Workflow
As we can see in the workflow diagram in Fig. 3, the
Raspberry Pi uses sensors to continuously collect data, including
temperature, humidity, sound, and weight. This data is send to
the server in frequent cycles and directly analyzed. Once the
analysis is completed, the determined sentiment result is passed
to the server for storage. If no action is required during one cycle,
i.e., when no activity is recorded, the next cycle begins
automatically. Once the Raspberry Pi determines that the baby
is in an abnormal state (i.e. crying or bed wetting), it
immediately starts the soothing mode to pacify the baby by
playing music and by gently shaking the crib. In addition, the
server sends a message to the mobile
Application installed on
Hardware
Sensor
Sensing
Raw data
Raspberry Pi
Raw data processing module
Action module
Setting
module
WIFI module
Other electronic equipment
ON and
OFF
switch
Settings
Data storage
Server
Response to
request
Send Request
Mobile
Terminal
App
Data
request
module
Message
receive
module
Remote
control
module
Data flow
Control flow
Figure 2. Architecture of Smart Crib System
Dostları ilə paylaş: