Driving License Corroborator Using Internet of Things and Data Mining Techniques: Analytical Essay

Abstract

Internet of Things allows objects to be sensed and controlled remotely across existing network infrastructure, creating opportunities for more direct integration between the physical world and computer-based systems resulting in improved efficiency, accuracy and economic benefit. Using IoT in vehicles could increase safety with a higher counter puncture level for accidents. Teenagers drive less than all but the oldest people, but their number of crashes and crash extinctions are indecorously high. The fatal crash rate per mile driven by 15 to 18-year-old people is nearly three times the rate for drivers with age 20. In order to disannul this fettle and to ensure the security of the vehicles, Driving License corroborator, the proposed model can be employed in such a way that the device assures the age of chauffeur (Driver) is above 18 with certain authorization steps.

Keywords: Internet of Things, Vehicle security, Driving License Corroborator, Authorization.

1 Introduction

The Internet of Things allows objects to be sensed and controlled remotely across existing network infrastructure, creating opportunities for more direct integration between the physical world and computer-based systems resulting in improved efficiency, accuracy and economic benefit.

The advancement in the field of transportation leads to the increase in the number of road users. Yet, the threat of road accidents is fast increasing in parallel to the count of vehicles on road. One of the causes for road accidents is the unlicensed drivers on road. According to industry reports, India accounts for 12.5 per cent (over 1, 45,000 fatalities a year) of global road accidents, with one road accident occurring every four minutes. Alarmingly, 72 percent victims involved in such road mishaps are between the age group of 15-44 years with speeding, reckless and drunken driving being the top reason accounting for 1.5 percent of road traffic accidents and 4.6 percent of fatalities. According to data from the Ministry of Road Transport and Highways, 10,622 persons under the age of 18 lost their lives to road crashes in India, accounting for 29 deaths on average every day. The proposed system helps to avoid such accidents, by primarily assuring that the age of the driver is above 18 and has been issued a driving license. Further, a vehicle with an identity verification system helps to get rid of misuse of vehicles [6].

2 Literature review

2.1 Fraud prevention

• Katarzyna Bobkowska et al. [1] proposed fingerprints and iris require cooperation from the traveler to provide his biometric data to the inspection terminal.
• Common biometrics has different challenges that can affect their performance in identifying individuals:
• Presence of glasses adversely affects using retina as biometric.
• Low light and movement when scanning an individual’s iris reduce the effectiveness of this biometric.
• Scarring, skin changes due to age and jewels can cause problems with hand geometry biometric.

2.2 Biometric Authentication based Vehicular Safety

• Elahi A et al. [2] proposed that biometric systems are present-time and approaches like fingerprint (biometric) recognition, iris recognition and facial recognition are becoming popular.
• Of these, fingerprint (biometric) recognition and detection systems are easy to deploy, sophisticated and persons can be identified without their knowledge.
• Biometric identification-based security systems are considered to be the most secure especially due to their ability to identify people with minimal ambiguity.
• It uses a fingerprint detection and recognition system that identifies and verifies a person automatically by extracting unique features from an image.

3 Motivation

The advancement in the field of transportation leads to the increase in the number of road users. Yet, the threat of road accidents is fast increasing in parallel to the count of vehicles on road. One of the causes for road accidents is unlicensed drivers on road. The rate of accidents caused by unlicensed drivers on road is found to be increasing as days go by. Another scary issue is that drivers who are not eligible to hold a license drive the vehicle is 6 out of 10.

The traditional system suffers from the following shortcomings:

• Transient storage facility is used to store user details that become non-permanent [5].
• Identity verification using cameras [1] is costly and the social acceptance among the people is medium-low.
• Iris recognition fails in case of less luminous intensity [1].

This motivated us to make the proposed system to overcome this problem.

4. Proposed model

The idea is to use the fingerprint sensors in the vehicles for authentication [3]. The fingerprint sensor is mounted on the fenders of the car which constantly detects the presence of a template and the license number in the sensor. The Arduino UNO with an inbuilt Atmel microcontroller is used to read the data from the sensor.

The proposed system uses four modules:

• 4.1 Arduino model setup.
• 4.2 Cloud storage structure for efficient storage and retrieval.
• 4.3 Preprocessing of user data.
• 4.4 Authorization using KNN classification algorithm.

4.1 Arduino Model Setup

The Arduino Uno is a open source microcontroller development board. The Uno differs from all preceding boards in that it does not use the FTDI USB-to-serial driver chip. Instead, it features the Atmega8U2 programmed as a USB-to-serial converter. Figure 1 shows the block diagram of the Arduino Model Setup.

Figure 1 Arduino model setup block diagram

In the proposed system, GSM module (connected to Arduino) is used to send license and insurance expire notification details to the phone via email. Relay is connected to Arduino with GSM module. Two motors are fixed with relay. Fingerprint sensor is placed on the board and LCD is mounted on the top of board to display the license details that the user has entered. Finally power supply will be given. Keypad is placed in the bottom of board that consists of five buttons that includes:

• License details.
• Insurance details.
• Entering the number randomly in ascending order.
• Entering the number randomly in palindrome.

4.2 Cloud Storage Structure For Efficient Storage And Retrieval

ThingSpeak is an open-source Internet of Things (IoT) application and API which can be used to store and retrieve data from things using the HTTP and MQTT protocol over the Internet or via a Local Area Network. ThingSpeak enables the creation of sensor logging applications, location tracking applications, and a social network of things with status updates. Figure 2 represents the process of retrieving data from the cloud storage.

Figure 2 Retrieval of data from cloud

The database containing the driver details is stored in the Thingspeak cloud [4]. When a person attempts authentication, the match for the fingerprint is searched in the cloud database and the action is categorized.

4.3 Preprocessing Of User Data

As KNN algorithm is very sensitive to noisy data, the data has to be preprocessed and smoothed out from noise. The steps in preprocessing of fingerprint are as follows:

• Grayscale transformation – The fingerprint returned by the fingerprint sensor has to be converted into grayscale. It assists with decreasing the space utilized for recording information remembered for the picture.
• Normalization – The standardization activity makes it conceivable to build the differentiations in the picture by making a specific change in accordance with the dim levels spoke to in every pixel without influencing the helpful data remembered for the unique finger impression picture.
• Segmentation – This activity means to limit the clamor present in the picture just as to focus the helpful data that will be utilized in the correlation period of the various marks, so we take out the loud territories just as the futile edges of the picture since they can lead us to make bogus examinations.
• Directional map – It present the direction of each streak contained in the unique mark. For the situation where the info picture is too boisterous we can have bogus outcomes when recognizing the neighborhood bearings of the pixels. Along these lines we can take note of the nearness of vertical lines at the directional guide levels that have no heading.
• Frequency map – It comprises of evaluating the nearby recurrence of the striations in every pixel.

4.4 Authorization Using KNN Classification Algorithm

Although methods like SVM and Adaboost algorithms are proved to be more accurate than KNN classifier, KNN classifier has a faster execution time and is dominant than SVM.

The steps of KNN classifier algorithm are as follows:

• Input: All samples of fingerprint.
• Output: Authentication categorized into authorized and unauthorized.
• Determine the factor K = number of nearest neighbors.
• Calculate the distance between given sample and training samples.
• Sort the distances and determine the nearest neighbor based on minimum distance.
• The majority of the nearest neighbor is used as prediction value.

5 Experimentation and results

This section presents the experiment results after testing the proposed system.

The Arduino model setup has been done. Figure 3 shows the experimentation of Arduino model setup.

Figure 3 Arduino Model Setup

The thingspeak cloud environment is set and figure 4 shows the home page of ThingSpeak portal after login.

Figure 4 Setting up of Cloud environment

The fingerprint sensor scans the fingerprint and preprocesses the input fingerprint. After feeding the training data into the ThingSpeak platform, the figure 5 shows the set of all available fingerprints.

[image: Image result for 30 human thumb impression images]

Figure 5 Human thumb images

All the images are converted into gray scale images and cropped to standard 256×256 sizes. Figure 6 shows the image of a preprocessed fingerprint.

Figure 6 Preprocessed thumb image

Driving License Corroborator initially prompts the user to enroll them providing license id and fingerprint. The enrolled data is stored in the inbuilt memory. Figure 7 shows the process of enrolling using license id and fingerprint. When the user wants to use the vehicle, he has to enter the license id and authenticate with fingerprint. Figure 8 shows the Driving License Corroborator prototype.

Figure 7 Driving License Corroborator prototype

Figure 8 Driving License Corroborator – Prototype

6 Conclusion

In this paper, we have proposed a novel approach that can be used to achieve better security of vehicle and prevent individuals without driving license from driving vehicles. It includes the efficient usage of fingerprint recognition for authentication of the drivers driving the vehicle. It helps to improvise the existing manual vehicle ignition system with certain additional features. As the outcome of authentication is associated with the vehicle ignition system, no manpower required for checking the user details.

7 References

1. Incorporating iris, fingerprint and face biometric for fraud prevention in e-passports using fuzzy vault – Katarzyna Bobkowska; Khaled Nagaty; Marek Przyborski.
2. Biometric Security-based Vehicular Safety System Nikhita Bhole, Leena Joag, Priyal Varma, Neha Utekar & Dipti Karani – International Journal of Interdisciplinary Research (IJIR) Vol-3, Issue-4, 2017.
3. Authentication-based systematic driving license issuing system – N. Ramakumar, P. S. N. Reddy, R. N. Naik and S. A. K. Jilani, ‘Authentication based systematic driving license issuing system,’ 2017 International Conference on Intelligent Computing and Control Systems (ICICCS).
4. Vehicle And License Authentication Using Finger Print – S. Prema; Mohamed Riyas V. S. Deen; Murali V. P. Krishna; S. Praveen 2019 5th International Conference on Advanced Computing & Communication Systems (ICACCS).
5. Authentication in Transient Storage Device Attachments – Donald Rich Computer Year: 2007.
6. Device with identity verification – Apply in car driving as an example Yu-Shuang Huang; Chi-Hao Lung J. Angeline Rubella; M. Suganya; K. Senathipathi; B. Santhosh Kumar; K. R. Gowdham; M. Ranjithkumar 2012 Fourth International Conference on Advanced Computing (ICoAC) Year: 2012.