OCR and RFID Enabled Vehicle Identification And

OCR and RFID enabled Vehicle Identification and

Parking Allocation System

Asmita Jondhale1 Aarti Thakare2 Samadhan Sonavane3

Computer Engineering

Sandip Foundation’s SITRC, Nasik, India

{, , }

Abstract: The available parking management systems require human efforts for recording entries in excel sheets. For huge parking it is a bit hectic to keep track of the information. Use of RFID technology reduces human efforts and OCR enabled system will provide an automated system for parking management. And also it will provide access control by use of boom barriers. It will be an efficient system for huge parking. Parking can also be classified on the basis of vehicles. Like if trucks are being parked in parking then it will be called commercial. If only staff vehicles are parked then it will be private and VIP for important people's vehicles. Vehicles cannot access parking area without permission. For companies it is an effortless and the best system. Having ease of access and control. It also contains cameras for surveillance. System utilizes two strong technologies RFID and OCR.

Keywords: Boom barrier, Number Plate, OCR, Parking, RFID, Reader, Tag, Template, Wallmart

INTRODUCTION

RFID technology stands for Radio Frequency Identification. This technology is used to track things and things include all possible things that we can think of. From a new born baby to a giant elephant. RFID readers are used to read tag values from the thing on which it is tagged. It includes RFID reader and RFID tags RFID writers etc. This technology is older than bar codes. For the first time it was used in World War II for airplane identification. In 1994, RFID technology was used by all rail cars in United States for identification. RFID is an old technology. It is better than bar codes. As we have two types of tags. Active tags and passive tags. Passive tags are having unlimited lifetime and they are low cost. Does not re-quire its own power supply. Whereas active tags are costly as they require power. Still it has not been used at greater extent in industry, although it has been known for a long time. The reason is its cost and lack of standardization among the manufacturing companies. It took a long time to be used widely. Wireless storage and automatic retrieval of data are the main benefits of using RFID. [6] This technology is an improvement over the bar code system and conventional identification, tracking, stocking of objects. OCR (Optical Character Recognition) is a technology used to recognize characters. It captures the image of Number Plate and performs image extraction and Template matching to identify the image. It includes preprocessing, character segmentation and character recognition. This technology is efficiently used in abroad. It requires a specific format of number plates. It is also available in different formats. To recognize English and other language number plates.[2] For OCR, the source is a scanned copy or a snapshot or photo. And destination is ASCII/UTF text, HTML/XML, TEX.

LITERATURE SURVEY

A lot of work had been done in past on RFID. It is used in many applications. Like shopping, transportation, hospitals, entry control, access control etc. let us review what work has been done. RFID used in shopping mall (wall mart), in this, the RFID tags are mounted on the shopping carts. Shoppers can be guided electronically to and desired products that are tagged with RFID tags. It also helps to track the shopper's location by RFID readers located at different places in the shopping mall. RFID tags also verify the items purchased by shopper. And at the counter it communicates with the billing system and also generates the bill.[5]

RFID used in hospitals. The paper which proposed this system said. In hospitals RFID tags can be used to identify patients, equipments, new born babies. The patients wear RFID tags on their wrist. Hence can be tracked. It gives automated patient identification. This can be easily implemented in hospitals, healthcare systems, fitness centers etc. it reduces capital required in these areas extensively. Even to keep track of equipments in the hospital RFID can be used. RFID tags can be mounted on almost anything. From a small baby to a big bus. RFID is highly secure.[7]

RFID in smart parking system, this system gives an automated parking system as we said earlier. Entry in parking, vehicle identification, Parking allocation, exit everything is automated. Using RFID tags on vehicles and readers at the entrance. The system is very much feasible for parking management. With reduced man power and use of technology. RFID makes system more feasible due to its significance in past works. And OCR will prove to be a good technology for the unregistered vehicles. This can also be said as the disadvantage of previously implemented system.[7]

OCR technology has been used in Abroad for Number Plate recognition for secure vehicle identifications. And also it is in use in forensic investigation applications. This technology for Number Plate recognition requires a fixed format of Number Plate to compare it with input. Or it records the changes present in both and processes it. When the vehicle will be coming for parking, at entry point the reader will read the RFID tag and will verify if the entry is valid. If yes, then it will signal boom barriers to open and the vehicle will be allowed to get in.[3][2]

In case if the vehicle is not regular and came for the first time OCR will perform number plate recognition and will store that information in database. And on exit of that vehicle entry will be deleted from the database. If that vehicle is supposed to come every day then it can be registered in regular list of vehicles in database. [1]

ARCHITECTURE OF PROPOSED SYSTEM

1) For vehicle identification and parking allocation process the following architecture is used.

Fig 1. Vehicle identification and parking allocation

In above architecture it is shown that if the vehicle is registered then input of tag will be given to RFID reader, else Number Plate as input will be given to OCR. RFID reader will verify and store the date and time of check-in for that vehicle in database and allows check-in. While OCR stores number plate recognition output information in database. At the exit point the vehicle is identified again. If it tagged with RFID tag then reader reads it and stores the date and time of check-out for that vehicle tag in database. And if tag is not present then OCR identifies number plate and deletes the entry of that vehicle from database. And vehicle is allowed to check-out. The working flowchart is as shown.

Fig.2 Working Flowchart

MATHMATICAL MODEL

RFID

• Check-in:

Consider that there is a RFID reader at the entry point and it reads the RFID tags on vehicles.

Let us say that receiver obtains value 'X' for every input tag.

Let 'A' be the set having all 'n' values of vehicles coming at an instance.

Now, A={X1, X2, X3……. Xn}

When we retrieve values, X is assigned with X->ID of tag.

Let 'B' be the set having entries for all registered vehicles.

After matching 'X' with value in database it is stored in 'B'.

Now, B= {B1, B2, B3…….. Bn}

If not registered then store it in 'U'.

Along with checking we assign an area for parking to vehicles. Say, 'D'.

Now, B= {D1, D2, D3……..Dn}

Each 'X' in B contains Di value.

B={X1, X2, X3…………….Xn}

Therefore, X1= {D1}, X2= {D2}, X3= {D3}, Xn= {Dn}

• Check allocation of parking space:

Let 'E' be the area where vehicle is parked. Compare 'E' with 'D'.

F=X1, X2, X3……………...Xn

X1= {E1}, X2= {E2}……., Xn= {En}

Compare 'E' with 'D' of Xi in B { }.

If E=D then no changes to make, else replace D with E.

Following operations are carried out to up-date sets and get resultant set.

1. P1=B intersection F

1. P2=F-B

1. P=P1 union P2

• check-out:

Read 'X' value of vehicle which is going to exit and store it in set 'L'.

L=X1,X2, X3…….Xn

U=B-L

U is the updated set.

OCR

Let an input snapshot of number plate be defined by a 2D function f(x; y), where x, y are coordinates and f is intensity of light at that point.

1. Edge detection :

Periodical convolution of 'f' with specific matrix 'm' to detect various types of edges. Therefore,

Here, m[x, y] represents element at xth row and yth Coolum.

1. Convolution matrices

It defines each image operation. Following Fig 3 shows how the affected pixel is found by using its surrounding pixels.

The pixel represented by y in destination image is affected by X0.....X8.

The formulae are as given:

y=X0 mo+X1 m1+X2 m2+X3 m3+ X4 m4+X5 m5+X6 m6+X7 m7+X8 m8

(a) (b) (c)

Fig. 3 (a) source image (b) convolution matrix (c) destination image

1. Horizontal and vertical edge detection:

1. Sobel Edge detector:

ADVANTAGES

• The proposed system is very secure and re-quires least human efforts.

• Space in database is not wasted by use of OCR for unregistered vehicles.

• RFID technology overcomes the limitations of using Barcodes.

• The proposed system is reliable, easy to operate and effective for huge parking.

• RFID tags provides long range of reading. Performs reading within no time. [1][15][14]

DISADVANTAGES

• The proposed system is costly.

• Installation of proposed system requires RFID and OCR implementation knowledge.

• RFID signals can be interrupted by other signals if used at proper frequency.

• Cheap RFID tags may wear out after a du-ration of use.

• OCR requires a predefined template for matching which restricts the options for matching.[16][4][5][6]

ACKNOWLEDGEMENT

We are profoundly grateful to Prof. Samadhan Sonavane for his expert guidance and continuous encouragement throughout. And all the team members for their support.

CONCLUSION

The paper focuses mainly on the literature review and working of proposed system. And how we can combine and use effectively two technologies like RFID and OCR in one system. The proposed system overcomes the limitation of existing systems. And also shows an innovation in RFID usage. Thus, we can take benefit of two strong and in-demand technologies.

REFERENCES

[1] Mr. Amrut Ram Selokar , Prof. Shweta Jain, Automatic Number Plat Recognition System using A Fast Stroke-Based Method , Volume-1, Issue-7, April, 2014 .ion

[2] Rajshree Dhruw, Dharmendra Roy, Automatic Number Plate Recognition System, IJCSMC, Vol. 3, Issue. 7, July 2014

[3] Vandini Sharma1, Prakash C. Mathpal2, Akanksha Kaushik3, Automatic license plate recognition using optical character recognition and template matching on yellow color license plate, Vol. 3, Issue 5, May 2014.

[4] Dale R.Thompson, Senior Member, IEEE, Senior Member, IEEE, and Michael K. Daugherty Teaching RFID Information Systems Security,IEEE TRANSACTIONS ON EDUCATION, VOL. 57, NO.1,FEB. 2014.

[5] ADVANTAGES AND DISADVANTAGES OF THE USE OF DIGITAL PROCESSING IN EDUCATION

[6] Suen, C.Y.; Plamondon, R.; Tappert, A.; Thomassen, A.; Ward, J.R.; Yamamoto, K. (1987-05-29). "Future Challenges in Hand- writing and Computer Applications". 3rd International Symposium on Handwriting and Computer Applications, Montreal, May 29, 1987. Retrieved 2008-10-03.

[7] Zeydin pala, nihati inanc, utilizing RFID for smart parking application UDC, 65.011.56 2009.

[8] Sridhar Iyer IIT Bombay, RFID: Technology and Applications,

[9] B. A. Alsaify D. R Thompson, J. Di,”identifying UHF RFID tag using signal features at different tari durations”..IEEE Int. Conf. RFID, Orlando, FL, USA, Apr. 3–5, 2012, pp. 40–46.

[10] A. Juels, “RFID security and privacy: A research survey,” IEEE J. Sel. Areas Common. vol. 24, no. 2, pp. 381–394, Feb. 2006.

[11] D. Klair, K.-W. Chin, and R. Raad, “A survey and tutorial of rfid anti-collision protocols,” IEEE Common. Surveys and Tutorials, vol. 12, no. 3, Aug. 2010.

[12] K. Finkenzeller and D. Mller, RFID Handbook: Fundamentals and Applications in Contactless Smart Cards, Radio Frequency Identification and Near-Field Communication. Wiley, Aug. 2010.

[13] H. Lehpamer, RFID Design Principles. Artech House Publishers, Feb.2012

[14] J. Myung, W. Lee, and S. Jaideep, “Adaptive binary splitting for efficient rfid tag anti-collision,” IEEE Common. Letts. vol. 10, no. 3, pp. 144–146, 2006.

[15] Y. Maguire and R. Pappu, “An optimal q-algorithm for the iso 18000-6c rfid protocol,” IEEE Trans. Automation Science and Engineering, vol. 6, no. 1, pp. 16–24, 2009.

[16] E. Ilie-Zudora, Z. Kemnya, F. van Blommesteinb, L. Monostoria, and A. van der Meulen, “A survey of applications and requirements of unique identification systems and rfid techniques,” Computers in Industry, vol. 62, no. 3, p. 227C252, April 2011.

[17] Ondrej Martinsky, Burno university of technology, Algorithm and mathematical principles of Automatic number plate recognition systems.

1