Tracking people in smart cities for security purpose Essay

Tracking people in smart cities for security purpose

Ahmad Alkhodre, Abdullah N. Alharbi, Yasser K. Aljohani and Faris E. Allam

Faculty of Computer and Information Systems Islamic University of Madinah


Security has always remained a major challenge for organizations. With the advancement in technology, several solutions have been provided to track people in order to enhance security. This paper proposed a system to track people in public areas using modern techniques that can be implemented in smart cities. In this paper, the architecture of the new hybrid solution has been proposed that is based on two techniques i.e. cameras and sensors that work together for tracking people. When a person passes through the sensor, the sensor sends commands to the cameras to take pictures of the person and store it directly in the database with the date and time. For developing the system, waterfall methodology has been used. Furthermore, the implementation of the system has been carried out at the university level as a case study. The results of the study showed that proposed architecture can enhance security and solve challenges related to tracking people in smart cities.

Key words:

Tracking, Smart Cities, Waterfall Methodology, Sensors

1. Introduction

In the last few years, the issue of security has gained a lot of attention. The development of technology has resulted in increased security, but it also produced various issues such as system attacks, information stealing, security breaches, and cyber threats. Organizations have become vulnerable to attacks and always focus on leveling up security.

Organizations face various issues due to the lapses in the security system. There are various installed systems which utilize RFIDs but still faces difficulty in recognizing the intruder due to the fact that it can only instigate an alarm. Even in some smart cities, RFIDs have proved to show limited capability due to lack of storing the data and recognizing the intruder.

Although RFIDs are a great initiative for analyzing a person's activities and behavior in the organization, security cannot be compromised due to the lack of cameras. It is necessary for smart cities or agile cites to have a tracking system as a building block for achieving security goals. This paper proposes a hybrid system which not only detects the person but also stores information for further analysis.

The proposed hybrid system comprised of three major components which are as follows:


A couple of sensors used to determine tags that has passed through it, attached to an object with x and y for tracking the people moving by the passive and active tags, the sensors also send commands to the cameras to take pictures.


Fixed cameras in various places that connected to the sensors to capture pictures for the people who pass through the sensors and send the information to the database server to be stored.


A database server stores all the information that is captured by the cameras and the information readied by sensors. So, it can be later used to analyze for security.

As a case study, the proposed system will be implanted in the IU campus. Currently, the university gates have only sensors which make it vulnerable to potential attacks. With the implementation of this system, we will add cameras to cut down the unauthorized introduction of cars by analyzing the data from the snapshots taken from the cameras and filter them. Additionally, tags will be allotted to employees which will contain the information of each employee. The hybrid system will first take pictures through cameras for a person who will pass through the sensors and then all sensors which are distributed on campus will track him. Additionally, snapshots will also be taken to analyze further. This project aims to enhance the security of the premises and we hope that it would also result in increased security for smart cities too.

Fig. 1 System Architecture

2. Related Work

[1] propose a system that has the capability of tracking, counting, and detecting people that passes through a range of space using sensors called time-of-flight (ToF). In [2] author states that major difficulty for a system is to track an object when there are also other moving objects. The author proposed the system that tracks a human and divide the body into six parts i.e. left and right head, left and right arms, and left and right legs. This system employed a matching technique between parts of the same body. The system has fast reorganization using x and y but it requires a lot of cameras in an area and range of cameras is also lower. In [3] authors used a software system which supports a camera video to track people by tagging, identifying, head tracking and tracking multiple people at the same time. In [4] authors utilize an including framework dependent on face identification, right off the bat set following region and checking territory. At that point, authors utilize AdaBoost calculation to distinguish appearances to track them after location and judge. The system processes the video by observing picture at the passageways of open places, for example, research centers, classes, general stores, and so forth. In these photos, they utilize a region totally the general population and after that identification of a passing person is carried out. They utilize AdaBoost learning calculations to recognize faces. The pith of face detection is to order the location region into the face region and none-confront zone. As some non-facial zones are additionally characterized as face territories, so they must utilize skin recognition to enhance detection precision. They used the cam shift algorithm to track the faces after detecting them.

Tracking and people detection is an important viewpoint in video investigation and computer vision as well. In [5] authors examine the identification and tracking of people in video groupings caught from a stationary camera. A successful people recognition system calculation is proposed considering the bi-directional projection histogram of the grayscale two-out line differencing picture, proposing a successful strategy for various individual's divisions. A straightforward connected strategy, called directional closest neighbor coordination is produced for the following individuals. The weakness of this approach is that it just utilizes the projection data of the gray value as the identification foundation. Still, it has a higher performance which can capture between 2-6 objects per frame at the same time and has faster video stream analyzing. In [5] the author proposes a technique to recognize, perceive and track people using fixed cameras settled on a building. The technique comprises of two free stages. One is devoted to recognizing and track any moving item inside the picture outline using Particle Filter (PF). The other one is in control to dispose of any moving item that is anything except a human being. To play out the first errand, a Molecule Channel calculation is utilized, in such a way that it can play out the following of numerous articles. For the acknowledgment arranges a PCA (Essential Parts Investigation) strategy is connected to a few body parts (head, arms, and so forth).

While if the smart cities are considered, there are various characteristics which make a smart city. In [6] authors define the smart cities as those communities that have a cutting-edge urban idea which is fundamental for individuals to have a quality life. It is the reasonable perspective of a collection of different innovations to accomplish shrewd and economical practices. The authors propose a keen city considering the general methodology and the 3-C idea that characterizes the center character of the keen city.

In [7] smart cities are defined as the technologies like (IOT) Internet of Things, Mobile Robots, huge analyzing of data Human cluster and Cloud Computing with high quality to improve the life of citizens. In [8] authors determined the applications of the smart cities as utilized in relatively every aspect of our lives. These applications have been used to improve quality of life. In this investigation, a model application was made utilizing the proposed framework.

RFID gateway systems have got a lot of attention due to their huge capability in various settings. According to [9] a basic configuration of an RFID system for security purpose is to make a gateway that contains an antenna and identifier targets (tags) which contain information about the object that holds the tags to reduce the interference.

Tracking people groups' development without utilizing individuals GPS beacons is useful for security reasons [10]. It doesn't expect clients to wear any sensors or gadgets, However, the systems require costly equipment, for example, ultra-high-recurrence or infrared sensors, likewise have an issue of attack of security, for example, camera-based procedures, or need now and again support, for example, the substitution of batteries.

In [11] authors covered different algorithms that have been proposed to support crowd management. The architecture and algorithms used RFID systems and WSNs. It makes a lot of services available for the pilgrims, including emergency services, health monitoring, and several location-based services. Also, his use of an RFID tag would also help authorities to track and identify pilgrims.

In order to detect humans, it is necessary to understand social behavior. According to [14] social behavior consists of a set of interactions among individuals of same species. There are various human social behaviors such as shaking hands, religious rituals, sharing meal, teaching, and disciplining a child. According to another research [15] which considered geo social crowd sensing platform that measurably evaluates and portrays the minimal way how socio-specialized asset accessibility differs in space and time such as populace thickness and district essentialness. While another research study [16] claims that access to the correct data at the perfect time permits natives, specialist organizations and city government to settle on better choices that outcome in expanded personal satisfaction for urban occupants and generally speaking manageability of the city. Social behavior is characterized as conduct that impacts or affected by individuals from similar species [17]. While another research study [19] suggested that social behavior comes in numerous structures such as squinting, eating, moving, shooting, and warning. Social behavior is influenced by the settings in which an individual grow [19 20]. Therefore, it is necessary to understand the behavior for designing the smart cities as when a smart city is designed, we focus on a desirable lifestyle for a citizen. It is the behavior of a citizen which shape its attitude towards smart cities [18].

All the suggested solutions in these case studies are very useful and convincing that sensors can be used to track people. But these systems also have some drawbacks too as mentioned with the analysis of each study. We want to first research thoroughly and then provide such a solution which not only fills the need for smart cities but also enhance the security of premises.

3. Proposed Architecture

3.1 Software Development Methodology

In this project, we used waterfall model. It is defined as "some basic tasks, which are carried out in a sequence of requirements definition, architecture design, detailed design, implementation, component verification, integration verification and requirements validation. Each task results in documents or other artifacts that are used as specifications for the next task, e.g. the detailed design specifications form the basis for implementation task. In the "ideal" form, one task should be completed before the next starts. There are however variants with overlapping tasks, and these are probably used more in reality. Implementation starts when some of the detailed design is ready [12].

Fig. 2 Software Development Methodology

3.2 Tools Used

1. ArgoUML: Open source-modelling tool to draw UML diagrams.

2. Visio 2016: Diagramming and vector graphics application.

3. Visual Studio: A development tool which makes application development easier for any platform & language.

3.3 System Analysis

1. Adding, uploading and deleting employee data.

2. Preventing unauthorized access.

3. Analyze employee behavior.

4. Modify location.

5. Monitor all camera across the campus.

3.4 Requirement Analysis

3.4.1 Functional Requirements

1. Analyzing the data.

2. Ability to Store data in a database.

3. RFID reader recording the data by detecting the Tags attached to an employee.

4. Tracking people.

5. Take a picture.

6. The ability to simulate the people's paths.

7. Ability to collect social information.

3.4.2 Non-functional Requirements

1. Accuracy

2. High security

3.5 Requirements Specification

3.5.1 User Requirements

1. Organization admin: The admin should have training on the system to be able to access to the data.

2. General Admin: Able to give help to the organization admin and give permission to employee, student, etc.

3.5.2 System Requirement

1. Desktop: i7 intel processor, 8GB RAM and 1TB hard disk.

2. UHF: RFID sensor.

3. Camera: To take a picture of person.

4. Arduino Uno board: to connect the camera with the system.

5. Database: SQL server management studio to creating the database.

6. Visual studio: To create the interfaces.

7. Operating system: Windows 10 Home.

8. Arduino software: upload the codes into the Arduino board.

3.6 UML Models

UML is Unified Modeling Language which provide the development community with a stable and common design language that could be used to develop and build computer applications [13]. The proposed project used four types of UML models which are as follows:

1. Use case diagram: "A use case illustrates a unit of functionality provided by the system. The main purpose of the use-case diagram is to help development teams visualize the functional requirements of a system, including the relationship of "actors" (human beings who will interact with the system) to essential processes, as well as the relationships among different use cases [13].

2. Activity diagram: "Activity diagrams show the procedural flow of control between two or more class objects while processing an activity "[13].

3. Entity Relationships Diagram: "is used for describing data and the relationship between different entities in a database. It creates a visual map outlining process requirements and detailing connections among entities and their attributes".

4. Sequence Diagram: A sequence diagram show parallel vertical lines for different processes or objects and, as horizontal arrows, the messages exchanged between them, in the order in which they occur.

3.6.1 Use Case Diagram

Fig. 3 Organization Admin Use Case

Fig. 4 System Work Use Case

3.6.2 Activity Diagram

Fig. 5 Add New Activity Diagram

Fig. 6 Delete Tag Activity Diagram

Fig. 7 Upload Tag Activity Diagram

Fig. 8 Search Tag Activity Diagram

Fig. 9 Add New User Activity Diagram

Fig. 10 Update User Activity Diagram

Fig. 11 User List Activity Diagram

Fig. 12 Delete User Activity Diagram

Fig. 13 Login Activity Diagram

Fig. 14 Logout Activity Diagram

3.6.3 Entity Relationship Diagram

Fig. 15 Entity Relationship Diagram

3.6.4 Sequence Diagram

Fig. 16 System Working Sequence Diagram

4. System Implementation & Testing

4.1 Database Implementation

The database implementation is done by SQL server management studio. Our database tables are as follows:

Admin Table:

Admin ID: Unique for admin

Name: Identify each admin user

Password: Security authentication

Fig. 17 Admin Table

Camera Table:

Camera_ID: No. for each camera

Place_ID: No. for each place

Capture_ID: Unique No. for each captured picture from the camera

Name: Name for each camera taken

Fig. 18 Camera Table

Capture Table:

Capture_ID: Unique No. for each captured picture

CapDate: Current time and date at time of capturing

Fig. 19 Capture Table

Place Table:

Place_ID: Unique No. for each place

Name: Name for each place placed on sensor

Fig. 19 Place Table

Sensor Table:

Staff_ID: Unique No. for each staff member

Name: Name of staff member

Dept_Type: Department of each staff member

Tag_ID: Unique No. for each tag holder

Fig. 20 Staff Table

Tag Table:

Tag_ID: Unique No. for each tag holder

Staff_ID: No. for each staff member

Capture_ID: Unique No. for each captured picture from the camera

Name: Unique name for each tag holder

Fig. 21 Tag Table

4.2 Interface Implementation

1. Login page: The login page allows the user to login with username and Password and to reset the password.

2. The menu page: Contains four main options and the logout button.

3. Add equipment page: To add the equipment of the system to a specific location.

4. Add user page: To add users to the system.

5. Monitor page: To monitoring the users and simulate the people’s path.

6. Cam page: For looking about a specific user by date and time.

7. User data page: To show all the information about the selected user.

8. View location page: To show all the locations in the system.

9. RFID page: To connect the RFID with the desk application.

10. View user page: To show and delete the users in the database

4.3 System Testing

Testing process is used to evaluate the system to specify requirements and to determine the gaps, errors in the system. In this project, we implement more than 75% of the project that is starting with passing a tag to the (RFID) sensor in order to be captured and then stored in the database with date and time in order to analyze. However, we connected the camera to the sensor to take an immediate picture of the passing object while the (RFID) is sensing the tag. The date and time helped us for tracking people and draw a path for their movement inside the campus.

5. Conclusion

Smart cities are a designation that is given to city which integrates data and correspondence innovations (ICT) to improve the quality and execution. This project aims to provide a new hybrid solution for smart cities that is combination of cameras and RFID sensor in order to level up the security. With the help of this system, the security for smart cities will increase.


[1] Rudolf, T., Martin, S., Adriano, Z., & Andreas, H. (2008). People Detection and Tracking with TOF Sensor. IEEE Fifth International Conference on Advanced Video and Signal Based Surveillance (pp. 356-361). IEEE.

[2] Zui, Z., Hatice, G., & Massimo, P. (2007). Tracking People in Crowds by a Part Matching Approach. IEEE Fifth International Conference on Advanced Video and Signal Based Surveillance (pp. 88-95). IEEE.


[4] Huajie, W., Limin, M., & Jinxue, P. (2013). Research of People Counting Based on Face. 3rd International Conference on Computer Science and Network Technology (pp. 525-528). Hangzhou, China: IEEE.

[5] Honglian, M., Halcyon, L., & Mingxiu, Z. (2008",). A Real-time Effective System for Tracking Passing People. Proceedings of the 7th World Congress on Intelligent Control and Automation (pp. 6173-6177). Chongqing, China: IEEE.

[6] Nallapaneni, K. M., Sonali, G., & Pradeep, M. K. (28-30 March 2018). Smart cities in India: Features, policies, current status, and challenges. 2018 Technologies for Smart-City Energy Security and Power (ICSESP). Bhubaneswar, India: IEEE

[7] Somayya, M., & R., R. (17-19 Feb. 2015). 100 New smart cities (India's smart vision). 2015 5th National Symposium on Information Technology: Towards New Smart World (NSITNSW). Riyadh, Saudi Arabia: IEEE

[8] Uğur, Ö., Aslihan, A., Menekşe, İ., & Enis, K. (2017). An augmented reality application for smart campus urbanization: MSKU campus prototype. 2017 5th International Istanbul Smart Grid and Cities Congress and Fair (ICSG). Istanbul, Turkey: IEEE

[9] Leandro, T. d., Maicon, C. E., & Luciano, L. P. (2012). Ultra high-frequency RFID gateway system for identification of metallic equipment. 2012 8th International Caribbean Conference on Devices, Circuits and Systems (ICCDCS). Playa del Carmen, Mexico: IEEE

[10] Wenjie, R., Quan, Z. S., Lina, Y., Xue, L., Nickolas, J. F., & Lei, Y. (2018). Device-free human localization and tracking with UHF passive RFID tags: A data-driven approach. Journal of Network and Computer Applications, (pp. 78-96).

[11] Tufail, A. (2018). Pilgrim Tracking and Location Based Services Using RFID and Wireless Sensor Networks. IJCSNS International Journal of Computer Science and Network Security, 112-119.

[12] S. Pathak and P. Saxena, "Hybrid Methodology Involving Scrum and Waterfall Model towards the Software Project Development in Academic Knowledge Centers", International Journal of Evaluation and Research in Education (IJERE), vol. 1, no. 1, 2012. Available: 10.11591/ijere.v1i1.456.

[13] Bell, D. (2003). UML basics: An introduction to the Unified Modeling Language: The Rational Edge

[14] Hamilton, W. D. (1964). The genetical evolution of social behaviour. International Journal of Theoretical Biology 7, 1-16.

[15] Giuseppe , C., Luca , F., & Paolo , B. (2013, JUNE). Fostering ParticipAction in Smart Cities: A Geo-Social Crowdsensing Platform. IEEE, pp. 112-119.

[16] Nasrin, K., Ali, M., & Mo, M. (2013). Impacting Sustainable Behaviour and Planning in Smart City. International Journal of Sustainable Land Use and Urban Planning ISSN 1927-8845, 46-61.

[17] Ian Q. Whishaw, Bryan Kolb, in The Laboratory Rat (Second Edition), 2006

[18] R. Rummel, "SOCIAL BEHAVIOR AND INTERACTION", [Online]. Available: [Accessed: 21- Apr- 2019].

[19] Rothbart, Mary K.; Ahadi, Stephan A.; Hershey, Karen L. (1994). "Temperament and Social Behavior in Childhood". Merrill-Palmer Quarterly. 40 (1): 21–39. JSTOR 23087906.

[20] Sakamoto M., Nakajima T., Akioka S. (2014) A Methodology for Gamifying Smart Cities: Navigating Human Behavior and Attitude. In: Streitz N., Markopoulos P. (eds) Distributed, Ambient, and Pervasive Interactions. DAPI 2014. Lecture Notes in Computer Science, vol 8530. Springer, Cham.

How to cite this essay: