Automated Toll Fee Collection and Crime Detection #1 Pankajkumar Kandhare, #2 Shubham Makeshwar, #3 Suraj Raut,

Transcription

1 ISSN Automated Toll Fee Collection and Crime Detection #1 Pankajkumar Kandhare, #2 Shubham Makeshwar, #3 Suraj Raut, #4 Harshal Mitkari, #5 Prof.D.R.Anekar 1 kandharepankajkumar@gmail.com, 2 shubhammakeshwar23@gmail.com, 3 suraj.raut2151@gmail.com, 4 hsmitkari@gmail.com, 5 devanekar@gmail.com #12345 Department Of Information Technology, Pune University SAOE Pune-48,SPPU, India. ABSTRACT Developing a system to pay the toll automatically and to reduce the crimes happening in the transportation department. In this system camera is used for capturing the image of the vehicle number plate. The captured image would be converted into the text using Automatic Number Plate Recognition with pattern matching algorithm. Toll amount will be deducted from motorist E-wallet account after vehicle enters at the toll booth. Automatic License Plate Recognition system is a real-time embedded system which automatically recognizes the license plate of vehicles. This paper presents how to control on different crimes happening in transportation department and control on this crimes. Keywords: Pattern matching algorithm, Automatic Toll Collection. ARTICLE INFO Article History Received: 2 nd April 2017 Received in revised form : 2 nd April 2017 Accepted: 4 th April 2017 Published online : 11 th April 2017 I. INTRODUCTION Considering the present toll collection system where each vehicle has to stop and pay tolls. Suppose the manual toll collection system is very efficient then for one vehicle to stop and pay taxes total time taken is up to 1 minute per vehicle. Automatic toll collection aims to eliminate the delay on toll roads by collecting tolls electronically. Automatic toll collection determines whether the cars passing are in that program, alerts enforcers for those that are not, and electronically debits the accounts of registered car owners and check for crime records of the vehicles. There are many toll collection systems which collects toll electronically but no one gives us information of that vehicles. Many vehicles enter at toll plaza pay toll and leave toll but our system will check all the crime records about that vehicles. Automatic toll collection system was developed to track the crimes happening in transportation department. When any vehicle enters at the toll plaza we will capture an image of number plate of that vehicle after capturing an image we will convert this image into text format and search this number into crime records. If no crime records are found then we will pass that vehicle from toll plaza and if any vehicle found in crime records then we will inform nearest police station about that vehicle. After checking for crime records we will move for toll collection process. In toll collection process we have e-wallet option every time when vehicle owner enters at toll plaza money deducted from this e-wallet account and vehicle owner can also refill that e-wallet when e-wallet is on low balance. Automatic toll collection system checks all the records about stolen vehicles or crime records of all vehicles. II. LITERATURE SURVEY AUTOMATIC TOLL COLLECTION SYSTEM USING RFID : Automated Toll Collection System using RFID used for collecting toll automatically. In this we do the identification by using radio frequency. Every vehicle will hold an RFID tag. This tag is having unique identification number assigned. This will be assigned by Regional Transport Office or traffic government authority. In accordance with those number we will store, all basic information as well as the amount he has to paid in advance for the Toll Collection systems. Reader will be strategically placed at toll plaza at center. Whenever any vehicle passes the toll booth, the tax amount will be deducted from his prepaid balance. New balance will be updated. Incase if anyone has insufficient balance in his account, his updated balance will be negative one. To tackle this problem, we have camera on the way to capture the image of respective vehicles. Vehicles didn t have to stop in a queue, this 2017, IERJ All Rights Reserved Page 1

2 translates to reduce Traffic congestion at toll booth and helps in lower fuel consumption. This is one of the very important advantage of this system [6]. ELECTRONIC TOLL COLLECTION SYSTEM USING PASSIVE RFID TECHNOLOGY: Electronic toll collection (ETC) systems using radio frequency identification(rfid) technology. Research on Electronic toll collection has been around since 1992, during which RFID tags began to be widely used in vehicles to automate toll collection processes. The proposed RFID system uses tags that are on the windshields of vehicles, through which information embedded on the tags are read by RFID readers; The proposed system eliminates the need for motorists and authorities at toll booth to manually perform ticket payments and toll fee collections, respectively. Data information are also easily exchanged between the motorists and toll authorities, thereby enabling efficient toll collection by reduce traffic and eliminating possible human errors. AUTOMATIC TOLL GATE SYSTEM USING ADVANCED RFID AND GSM TECHNOLOGY: Electronic Toll Collection (ETC) systems around the world are implement by DSRC (Dedicated Short Range Communication) technology. The concept proposed is of automatic toll tax payment system and the amount transaction information is sends to the cell phone of the that motorists through the GSM modem technology. It is an innovative technology for expressway network automatic toll collection solution. In this paper, the frame composing and working flow of that system is describe in it and data information is also easily exchanged between the motorists and toll authorities, thereby enable a more easy toll collection by reducing of traffic and eliminate possible human errors. III. PROPOSED SYSTEM The proposed method is for providing a fast and automatic toll collection at the toll booth. Camera is used to capture image of the vehicle and the barrier are used here for open and close when the vehicle is entered or exit at the Toll booth. Then vehicle number will check in database The vehicle information is stored in the database will get check. Based on vehicle number the amount for vehicle will automatic transfer to the toll booth.and in crime detection,vehicle information will get check in stolen vehicle data,if that vehicle found stolen,then alert message will be send to nearest police station otherwise vehicle pay tax and pass from toll booth. The main purpose behind this is for creating Automatic Toll Collection System. This toll collection system uses pattern matching algorithm, which works on capture image and convert into binary image. Other than that, there is no need of tag, This system is at low cost However, this proposed system require changes in the existing toll booth. Fig.3.1 Proposed architecture 3.1 Working of the proposed architecture The proposed system makes sure that the traffic at the toll gates security is present. Every vehicle has unique format of number plate (eg.mh 12 GQ 1234). The pattern matching algorithm work on capture image of the number plate and identifies the vehicle. By that system we can identify stolen vehicles. Automatic number plate recognition has important role in many applications, like automatic toll collection Number plate recognize a vehicle s number from an image or images taken black and white, or infrared camera. It has a lot of techniques, such as image input, gray scaling, threshold, filter, segmentation, thinning, cropping, scaling, feature extraction & matching, output. The different type of the number plate cause challenges in the Detection of license plates. The number plate recognition system extracts a number plate from a capture image. The first step is to capture the image using a camera. Then capture image taken as input and pattern matching algorithm work on that image.in that gray scaling applied on image after that thresholding and filtering should be apply. The next stage is to extract number plate from the capture image, feature such as the boundary, the color, or the existence of the characters. The third stage is segmentation of the number plate and extracts those characters by taking their color information, labeling them, or matches their positions with templates. The final stage is to recognize those characters which are extracted by template matching. At that time, lots of people is needed in toll booth. In a single barrier we need a worker for collecting the cash and another worker to operate the barrier. All this avoid by making the toll booth automatic. In this system, detecting the vehicle at first and after that captures the image of the front view of the vehicle. In this system vehicle number plate is localized first and then characters are segmented. This system designed for gray scale images so that can detects the number plate regardless of its color. Template matching used for character to be recognizes. The vehicle number compared with the database of all the stolen vehicles so that information about the vehicle type & to collect toll tax according to that. If that is not a stolen vehicle then toll collection is reduced and the barrier is opened. If that same vehicle is return in specific time, it can be considered. 2017, IERJ All Rights Reserved Page 2

3 If a stolen vehicle found, then gate won t open and a message send to the nearby police station to take strict action. We can report about stolen vehicle by sending that vehicle number to a specified number. All this get updated in the database,and that database can be monitored by the police. That the vehicle number is updated in the data base and crime detection is improved. System is implemented on toll booth for its automatic toll collection and stolen vehicle detection. Pattern matching algorithm is used for identifying the vehicle number. 3.2 PATTERN MATCHING ALGORITHM Working of Pattern Matching Algorithm: Figure Pattern Matching Algorithm working Grayscale algorithm: Most digital images are comprised of three separate color channels: red channel, a green channel, and blue channel. Layering these channels on top of each other creates a full color image. Various color models have various channels.sometimes the channels are colors or values. All grayscale algorithms utilize the three-step process: Get the red, green, and blue values of a pixel. Use math to turn those numbers into a single value of gray. Replace the original red, green, and blue values by the new gray value When describing grayscale algorithms, I m going to focus on step 2 using math to turn color values into grayscale value. So, when you see a formula like: Gray = (a Red + a Green + a Blue) / 3 The simple thing to do is to average the R,G,B components into a single number. This operation will be done for each pixel. For example, if you have a pixel with values R=201, G=101, B=51, then the grayscale will be ( )/3 = 117. The average the three values into one number The average shows how bright the pixel is, ignoring hue: 0 = complete dark, 255=complete bright Thresholding: From a grayscale image, thresholding can be used to create binary images. Separate out regions of an image corresponding to objects which we have to analyze. This separation is based on the variation of intensity between the object pixels and the background pixels. The purpose of thresholding to extract those pixels from image which represent an object. The information is binary the pixels represent a range of intensities. Thus the objective of binarization to mark pixels which belong to true foreground regions with a single intensity and background regions with different intensities. It stores the intensities of the pixels in an array. The threshold is calculated from using total mean and variance. Based on this threshold value each pixel is set to either 1 or 0. i.e. foreground or background. Thus here the change of image takes place only once. The following formulas are used to calculate the total mean and variance. The pixels divide into two classes, C 1 with gray levels [1,...,t] and C 2 with gray levels [t+1,...,l]. probability distribution for the two classes : Also, means for the two classes Using Discriminant Analysis, Otsu define the between-class variance of the thresholded image as For bi-level thresholding, Otsu verified that the optimal threshold t* chosen so that the between class variance B is maximized; that is, 2017, IERJ All Rights Reserved Page 3

4 Filter Median: A non-linear filter changes image intensity mean value if the spatial noise distribution in the image not symmetrical within the window. Standard Median Filter (SMF) is such non linear filter. Variance intensities in the image is reduced by Median Filter. The novel filter processing principles are based on the adaptive median filtering. Adaptive median filtering works in a rectangular kernel area S x, y and increase the size of S x, y during filtering operation, depending on certain conditions listed below. If the filter does find that the pixel at (x, y) is noise in the center of kernel, the value of the pixel replaced by the median value in S x, y. Otherwise, the pixel gray level value remain the same. Consider the definition: Z min = min gray level value in S x, y Z max = max gray level value in S x, y Z med = median of gray level in S x, y Z x, y = gray level at coordinates (x, y) S max = max allowed size of S x, y The median filtering algorithm works in two levels, denoted level A and level B, as follows: Level A A1 = Z med Z min A2 = Z med Z min If, A1 > 0 AND A2 <0, go to level B Otherwise, increase the window size If window size S max, repeat level A Or else, output, Z x,y Level B B1, Z x,y Z min B2, Z x,y Z med If, B1 > 0 AND B2 < 0, output Z x,y Or else, output Z med. Every time the algorithm outputs value, the window, S x, y is moved to the next location in image. The algorithm then is reinitialized and applied to the pixels in the new position. AMF can achieve good results in suppressing noises of various densities. sometimes It changes its kernel maximum size in order to suit for various conditions. One way is using different kernel mean filters to process images and determine the AMF kernel of maximum size. Segmentation: Segmentation is the process of partitioning of digital image into multiple sets of pixels or segments. The goal of segmentation is to simplify and change the representation of an image into something that will more meaningful and easy to analyze. Image segmentation is typically used to locate boundries and object example lines, curves, etc. in images. More precisely, image segmentation is the process of giving label to every pixel in an image such that pixels are of same label share specific characteristics [11]. Image segmentation result is a set of segments that collectively cover the complete image. Each pixels in a region are same with respect to some characteristics, like as color, intensity, or texture Edge detection is a well-developed field on its own in image processing. Region boundaries and edges are very closely related to each others, since there is a sharp adjustment in intensity at that region boundaries. The edges identified by edge detection are often disconnected. Segmentation methods can also be applied to edges obtained from edge detectors. The developed an integrated method which segments edges into curved and straight edge segments for parts-based object recognition, based on a minimum description length criterion that was optimized by split-and-merge-like method with candidate breakpoints obtained from complementary junction cues to obtain such as points at which to consider partitions into different segments. Thinning: Thinning is a morphological operation.it is used to remove selected foreground pixels of binary images, such as opening or erosion. It can used for several applications. it is used to tidy up output of edge detectors by reducing all lines to single pixel thickness. Thinning is normally applied to binary image to produces another binary image to get output. 2017, IERJ All Rights Reserved Page 4

5 The thinning operation related to the hit-and-miss transform like other morphological operations, the behavior of the thinning operation is decided by structuring element. They can contain both ones and zeros. The thinning operation is related the hit-and-miss transform and is expressed quite simply in terms of it. The thinning of an image I by a structuring element J: where the subtraction is logical subtraction defined by the thinning operation is obtained by translating the origin of the structuring element to each possible pixel position in the image, and at each such position comparing it with the underlying image pixels. If the background and foreground pixels in the structuring element exactly match foreground and background pixels in the image, then the image pixel situated below the origin of the structuring element is set to background zero or else it is left unchanged. The structuring element choice decide under what situations a foreground pixel will be set to background, and hence it decides the application for the thinning operation. Cropping: Cropping means to the removing outer parts of an image to improve framing, accentuate subject matter or change aspect ratio. Depending on the application, this may be done on a physical photograph, artwork or film footage, or achieved digitally use by image editing software. Scaling: Scaling operator performs a geometric transformation which is used to resize the image. Shrinking of size of image known as subsampling, is done by replacement of a group of pixel values by a arbitrarily taken pixel value from within this group or by interpolating within image pixel values in a nearer neighborhoods. Image zooming achieved by pixel interpolation or by replication. Scaling is useful to change the visual appearance of an image, to alter the quantity of information stored in a scene representation a low-level preprocessor in multi-stage image processing chain which operates on features of a particular scale. Scaling is special case of affine transformation. IV. FUTURE SCOPE 1. Change in Travel Patterns 2. Public Transport Improvements 3. Traffic Flow Improvements 4. Congestion Reduction 5. Better Environment 6. Revenue Generation 7. Increased Safety V. ADVANTAGES 1. Automated toll collection system is very fast and efficient mode for collection of toll charges at the toll plazas. 2. The average number of vehicles waiting in the queue reduces and so the average waiting time is reduced. 3. Fuel saving -this results in gas saving for the elimination of acceleration and deceleration results in reduction of the operating cost of the vehicles. 4. Enhanced cash handing -- there is no cash transaction for the etc lane so cash handling is reduced so difficulties with cash handling is eliminated. Thus aid in enhanced audit control by centralizing user accounts. 5. Payment flexibility -- the patrons do not have to worry about searching for cash for the toll payment. Since the patrons set up account for etc usage it gives customers the flexibility of paying their toll bill with cash, check, or even credit cards. 6. The average number of vehicles waiting in the queue reduces and so the average waiting time is reduced. 7. Reduce time!!! 8. Accident reduction it is observed that there is reduction in the number of accident caused near the toll plazas due to considerable decrement in congestion around toll plazas. Reduced man power. VI. CONCLUSION The paper presents, how crime detection combined with automatic toll collection provides a very fast and time efficient. There is no cash transaction for this automatic toll collection system so cash handling is reduced so difficulties with cash handling and delay due to cash is eliminated. Automatic toll collection may become an increasing important instrument within the big bundle of measures for regional demand and traffic on highways this aid in enhanced audit control by centralizing user accounts. Automatic toll collection may become an increasing important instrument within the big bundle of measures for regional demand and traffic management. VII. ACKNOWLEDGEMENT The authors would like to thank prof. D. R. Anekar for their assistance and guidance in preparing this manuscript. 2017, IERJ All Rights Reserved Page 5

6 REFERENCES [1] Khadijah Kamarulazizi, Dr.Widad Ismail Electronic Toll Collection System Using Passive RFID Technology in Journal of Theoretical and Applied Information Technology [2] Pranoti Salunkhe, Poonam Malle, Kirti Datir, Jayshree Dukale, Automated Toll Collection System Using RFID in IOSR Volume 9, Issues 2(Jan Feb 2013) [14] Shinde, A. S., and M. R. Dhage. "A Survey: Sparse Traffic Grooming and RWA Assignment Schemes in Optical Network." International Journal of Computer Applications (2014). [15] PB Alappanavar et al, Location Based Augmented Reality, International Journal of Scientific & Engineering Research, Volume 4, Issue 5, May -2013, , ISSN [3] Kumar Rajeev Ranjan, Abhinav Sinha Automatic Vehicle Registration System for Tollbooths. [4] Priyanka Chhoriya, Govinda Paliwal, Poonam Badhan Image Processing Base Automation Toll Booth In Indian Condition, April [5] Romić, Krešimir, Irena Galić, and Alfonzo Baumgartner. "CHARACTER RECOGNITION BASED ON REGION PIXEL CONCENTRATION FOR LICENSE PLATE IDENTIFICATION." Tehnicki vjesnik/technical Gazette 19.2 (2012). [6] Shilpa G. Lathkar et al, Online Digital Advertising on Public Display, International Journal of Computer Applications, Publication date, Jan 2013, Vol.66, Issue 10. [7] Akshay S. Kyatam, Tracking and Scheduling of State Transport Bus using RFID, May-2015, International Journal Of Engineering And Computer Science Volume 4, Issue 5, pp [8] Aher, Madhuri, Poonam Pate, and Apurva Varade. "Analysis of Fractal Intraframe and Interframe Video Coding." International Journal of Global Technology Initiatives 3.1 (2014): B40-B46. [9] Kulkarni, Niraj, et al. "Multi-Agent System for Detecting and Blocking SQL Injection." International Journal of Computer Applications (2013). [10] Mohammed Kagalwala et al, Online Banking Security System Using OTP Encoded in QR-Code, March 2015, International Journal of Advanced Research in Computer Science and Software Engineering (IJARCSSE), Volume 5, Issue 3, ISSN(Online): X. [11] Bangare, Sunil L., et al. "Implementing Tumor Detection and Area Calculation in MRI Image of Human Brain Using Image Processing Techniques." International Journal of engineering Research and Applications 1.5: [12] Bangare, Sunil L., et al. "Reviewing Otsu s Method For Image Thresholding." International Journal of Applied Engineering Research 10.9 (2015): [13] Bangare, Sunil L., et al. Review and Design of Image Inpainting Technique using Novel Algorithm, May 2014, International Journal of Research in Information Technology, Volume 2, Issue 5,pp , IERJ All Rights Reserved Page 6