Low Cost Real-Time System Monitoring

Using Raspberry Pi

Huu-Quoc Nguyen, Ton Thi Kim Loan, Bui Dinh Mao and Eui-Nam Huh
Computer Engineering Department
Kyung Hee University
Yongin, South Korea
Email: {quoc, loanttk, mao.bui, johnhuh}@khu.ac.kr

Abstract—Nowadays, the Closed-Circuit Television (CCTV)

surveillance system is being utilized in order to keep peace and
provide security to people. There are several defects in the video
surveillance system, such as: picture is indistinct, anomalies can-
not be identified automatically, a lot of storage spaces are needed
to save the surveillance information, and prices remain relatively
high. This paper describes the design and implementation of a
low-cost system monitoring based on Raspberry Pi, a single board
computer which follows Motion Detection algorithm written in
Python as a default programming environment. In addition,
the system uses the motion detection algorithm to significantly
decrease storage usage and save investment costs. The algorithm
for motion detection is being implemented on Raspberry Pi, which
enables live streaming camera along with detection of motion. The
live video camera can be viewed from any web browser, even from
mobile in real-time.
Keywords—Video surveillance; Raspberry Pi; Motion Detection;
streaming
Fig. 1: System Architecture
I. I NTRODUCTION
Closed-circuit television monitoring system has now be-
come an indispensable device in todays society [1]. Supermar- Detection promotes an automated approach to video system
kets, factories, hospitals, hotels, schools, and companies are reaction, and does not require an individual to start and stop
having their own CCTV system for 24/7 monitoring. Instead recording sessions. So we can significantly decrease storage
of using the traditional wireless CCTV surveillance cameras, usage and save investment costs.
customers can now own their inexpensive security systems The rest of this paper is organized as follows. In Section
with the tiny super computer called Raspberry Pi [2]. IP II presents the system architecture. Section III presents the
cameras can serve better as they can send and receive data Motion Detection algorithm. The experiments and results are
via computer network and internet based on internet protocol. demonstrated in Section IV. Finally, conclusions and future
Also resolution clarity of IP cameras is far better than CCTV works are given in Section V.
cameras. A camera module connected to the Raspberry Pi
will records all the happenings in the monitored area and live
II. S YSTEM A RCHITECTURE
streaming can be viewed from any web browser, even from
mobile in real-time. The system architecture is shown in Figure 1. The camera
module connected to the Raspberry Pi board can be used to
There are many problems in the video surveillance system
take high definition video, as well as stills photographs.The
[3], such as: picture is indistinct, anomalies cannot be identified
setup Python script will automatically deliver video data
automatically and a lot of storage spaces are needed to save
streaming to cloud server. Users will be able to watch the
the surveillance information. Moreover, in recent years, Motion
video stream from cloud server on any device that has a web
Detection [4], [5] has attracted a great interest from computer
browser. This includes the IPad/iPhone and Android devices.
vision researchers due to its promising applications in many
areas, such as video surveillance, traffic monitoring or sign
language recognition. To overcome storage spaces issue, we A. Overview of the Raspberry Pi
apply the Motion Detection algorithm for live camera stream- The proposed system, uses Raspberry Pi Model B+ single
ing, this allows the system to analyze incoming images from board computer and offers these key features:
cameras, and recognize when movement occurs. And then,
the video system can collect and store the most importance • Broadcom BCM2835 SoC processor with 700MHz
items for the administrators review. Effective use of Motion ARM1176JZF-S core

978-1-4799-8993-5/15/$31.00 ©2015 IEEE 857 ICUFN 2015

• 512MB RAM Algorithm 1 Motion Detection algorithm
• Videocore 4 GPU supports up to 1920x1200 resolution 1: procedure M OTION D ETECTION
• 5Mpix Camera module capable of full HD video @ 30fps 2: calculate the average of a selected color in frame 1
• MicroSD card slot, 10/100Mbps Ethernet port, 4 x USB 3: wait X seconds
2.0 ports, HDMI, audio/video jack, GPIO header, mi- 4: calculate the average of a selected color in frame 2
croUSB power port, DSI and CSI ports 5: if abs(avgF rame1 − avgF rame2) > threshold then
• Dual step-down (buck) power supply for 3.3V and 1.8V 6: motion detected
During initial setup Raspberry Pi was configured as a
miniature desktop with USB configured camera module and an
external monitor or mobile for viewing the captured video. The
Raspberry Pi runs Raspbian OS and is programmed using GNU
Octave and Python, which is an open source. The Dynamic
host Protocol is obtained for Raspberry Pi, addressed IP is
fetched. After obtaining this the system can be configured and
controlled remotely.

B. Connecting the Camera

Fig. 3: Motion Detection Flowchart

Motion Detection algorithm also cant handle a rotating object

- an object that moves.
Fig. 2: Connection of Camera Module on Motion Detection flowchart is shown in Figure 3. If there is
Raspberry Pi board no motion detected, the program will not save the videos data.
Otherwise, if motion is detected, the current frame of detected
motion will be processed by Motion Detection algorithm. And
As shown in Figure 2, a 5MP camera module that capable then the system will be record full-hd videos, save them on
of 1080p video and still image, and it can connect to Raspberry the sd-card packed into MP4 container while the live-preview
Pi directly with CSI (Camera Serial Interface). And then boot continues.
the latest version of Raspbian and we are good to go with the
camera. Its a fixed focus 5MP sensor capable of 2592x1944 IV. E XPERIMENTS & R ESULTS
stills, but also 1080p30, 720p60 and 640x480p60/90.
Figure 4 presents the screenshots of the system. The control
Cost of the whole system is around $75. Includes the panel with control options to save stills and full-res MP4 file,
camera module costs, Raspberry Pi board and wifi adapter. capture a picture, timelapse, change the resolution of videos
So the system cost is less, compared to other alternatives. (Figure 4a). When movement occurs, the system will analyze
incoming images and store the most importance items. Go
III. M OTION D ETECTION A LGORITHM to the ”Download Videos and Images” link, we can view
a preview of the JPEG images and videos play smoothly
Motion detection works on the basis of frame differencing - (Figure 4b, 4c). We also can watch videos on mobile devices,
meaning comparing how pixels (usually blobs) change location it’s got a smooth set of controls, reliable performance, and a
after each frame. The method looks for a object change in the clear picture with no blips, glitches (Figure 4d).
image:
The latency remains fairly low at about 0.4 seconds for
The problem with these motion detection methods is that motion detection based on comparing an on screen timer and
neither detects slow moving objects, determined by the sensi- the video image of this system. While remotely viewing the
tivity of the threshold. But if the threshold is too sensitive, it 640x360 MJPEG image, the Raspberry Pi reports 67% CPU.
will detect things like shadows and changes in sunlight. The But it’s not overclocked.

858
TABLE I: Comparison of data storage between
with and without motion detection

Storage Space (GB/month)

Non-Motion Detection 236

Motion Detection 740

Table I presents a comparison of the storage space between

with and without motion detection. Storage space when using
motion detection is 236 GB/month, while the we take 740
GB/month if not using motion detection. The motion detection
algorithm would use help to minimize the recorded data storing
capacity. It means that we can save money for the storage fee.
(a)
V. C ONCLUSION & F UTURE W ORK
An approach for the surveillance monitoring system and
the Motion Detection algorithm to decrease storage usage
based on Raspberry Pi single board computer was proposed
in this paper. This is still far from being final version of the
monitoring system and there is lot for us to improve.
In future we plan to improve the Motion Detection algo-
rithm. Because the algorithm depends on threshold value, it’s
mean the performance of the algorithm can be enhanced by
considering certain conditions. If we have a good solutions
to get a good threshold value and then the algorithm can
(b) detect moving objects precisely, including slow moving or tiny
objects.

VI. ACKNOWLEDGMENTS
This research was supported by Next-Generation Informa-
tion Computing Development Program through the National
Research Foundation of Korea(NRF) funded by the Ministry
of Science, ICT & Future Planning (NRF-2010-0020725). This
research was supported by the MSIP(Ministry of Science,
ICT and Future Planning), Korea, under the ITRC(Information
Technology Research Center) support program (IITP-2015-
(H8501-15-1015) supervised by the IITP(Institute for Infor-
mation & communications Technology Promotion)). Eui-Nam
(c) Huh is the corresponding author.

R EFERENCES
[1] Yong-ik Yoon, Jee-ae Chun, Tracking System for mobile user Based on
CCTV. Information Networking (ICOIN), 2014 International Conference
on, Phuket, 10-12 Feb. 2014, pp. 374-378.
[2] Viren Pereira, Vandyk Amsdem Fernandes, Junieta Sequeira, Low Cost
Object Sorting Robotic Arm using Raspberry Pi. Global Humanitarian
Technology Conference - South Asia Satellite (GHTC-SAS), 2014 IEEE,
Trivandrum, 26-27 Sept. 2014, pp. 1-6.
[3] Yimamuaishan.Abudoulikemu, Yuanming Huang, Changqing, A Scalable
(d) Intelligent Service Model for Video Surveillance System Based on RTCP .
Signal Processing Systems (ICSPS), 2010 2nd International Conference
Fig. 4: System screenshots: (a) monitoring web interface; (b) on (Volume:3 ), Dalian, 5-7 July 2010, V3-346 - V3-349.
videos list after recorded; (c) preview videos real-time; (d) [4] C. Bahlmann, Y. Zhu, Y. Ramesh, M. Pellkofer, T. Koehle, A system for
traffic sign detection, tracking, and recognition using color, shape, and
interface on mobile devices motion information. IEEE Intelligent Vehicles Symposium, Proceed-
ings,2005, pp. 255-260.
[5] Adrienne Heinrich, Dmitry Znamenskiy, Jelte Peter Vink, Robust and
Sensitive Video Motion Detection for Sleep Analysis. Biomedical and
Health Informatics, IEEE Journal of (Volume:18 , Issue: 3 ) , 2168-2194,
20 September 2013, pp. 790-798.

859