Currency Based Internet Access System using ... - onlinepresent.org

9 downloads 0 Views 1MB Size Report
extracted ROI can be used with various Pattern recognition and Artificial ... required the currency note image by simple or normal flat scanner on fix dpi color with a ... neural network process is applied for matching to identify note value.
Advanced Science and Technology Letters Vol.147 (SMART DSC-2017), pp.433-440 http://dx.doi.org/10.14257/astl.2017.147.62

Currency Based Internet Access System using Artificial Neural Networks Bosubabu Sambana Assistant Professor, Dept.of CSE, Avanthi’s Research and Technological Academy-Bhogapuram, Jawaharlal Nehru Technological University-Kakinada, Andhra Pradesh, India. [email protected]

Abstract. We propose a new system which provides internet access by inserting categorizing the currency notes. Image processing technique to match the identified input required currency’s image and the reference image. For each currency a Region of Interest (ROI) is taken on existing environment. The extracted ROI can be used with various Pattern recognition and Artificial Neural Networks matching techniques. Once the two or more currencies notes are matched then internet based tracking system to detect matching technique is granted for accessing for their particular period of time. Initially we acquire required the currency note image by simple or normal flat scanner on fix dpi color with a particular size, the pixels level is set to obtain image using image processing technique. Few modern filters are applied to proposed extract denomination value of currency note. Then this value is compared with the input sign value to match original the currency note value. Then the internet based access is controlled by means of microcontroller to analyze all requirement fields and necessary actions. Key words: Flat Filters, Currency, ROI, Pattern Recognition, Neural Networks, Image Processing, Gray scale Images.

1 Introduction In this paper we scanned the different denomination notes at 150 dpi with 128x128 pixels. We extract denomination value from each note. A level is set for all images. By converting compliments, applying different filter i.e. sobel edge filter, average filter, laplacian filter, denomination value is extracted. The pattern recognition and neural network process is applied for matching to identify note value. Then this matching result is used as the input for providing access to the internet through different networks Single layer feed forward network: The simplest quite NN could be a Single Layer Perception (SLP) network, which consists of one layer of output nodes; the inputs square measure fed on to the outputs via a series of weights. The total of the merchandise of the weights and also the inputs is calculated in every node. During this single-layer feed forward NN, the network's inputs are directly connected to the

ISSN: 2287-1233 ASTL Copyright © 2017 SERSC

Advanced Science and Technology Letters Vol.147 (SMART DSC-2017)

output layer perceptions. Input nodes or units square measure connected to a node or multiple nodes within the Multilayer network: A multilayer feed forward NN is an interconnection of perception’s within which knowledge and calculations flow during a single direction, from the input file to the outputs. A multilayer feed forward NN consists of a layer of input units, one or a lot of layers of hidden units, and one output layer of units. A NN that has no hidden units is termed a Perception. The structure includes a layer of process units i.e. the hidden units additionally to the output units. These networks are referred to as feed forward as a result of the output from one layer of neurons feeds forward into consequent layer of neurons. Feed Forward neural network: A feed forward NN could be a biologically impressed Classification rule. It consists of an oversized variety of easy neuron-like process units organized in layers. Each unit during a layer is connected with all the units within the previous layer. During this network the data moves solely in one direction i.e. forwards: From the input nodes knowledge goes through the hidden nodes and to the output nodes. There are not any cycles or loops within the network. These associations aren't all equal; every connection could have a special strength or weight. The weights on these connections cipher the data of a network. Usually the units during a NN are referred to as nodes. Information enters at the inputs and passes through the network layer by layer till it arrives at the outputs. Throughout traditional operation there's no feedback between layers. This is often why they're referred to as feed forward neural networks. Feed forward networks will be created from differing kinds of units’ e.g. binary McCulloch- Pitts neurons the best example being the perception. Continuous neurons oft with colon activation are employed in the context of back propagation of error. Recurrent network: A repeated Neural Network (RNN) could be a category of NN wherever connections between units type a directed cycle. This creates an interior state of the network that permits it to exhibit dynamic temporal behavior. Not like feed forward NN, RNNs will use their internal memory to method capricious sequences of inputs. The human brain could be a RNN i.e. a network of neurons with feedback connections. It will learn several sequence process tasks or algorithms or programs that aren't learnable by ancient machine learning ways. A RNN includes a cyclic of its connections. Numerous characteristics of RNN are:   

All BNN square measure repeated mathematically, they implement energizing systems. Several varieties of coaching algorithms. Theoretical and sensible difficulties are prevented by sensible applications to date.

Pattern Recognition: The term pattern recognition encompasses a large vary of knowledge process issues of nice sensible significance, from speech recognition and

434

Copyright © 2017 SERSC

Advanced Science and Technology Letters Vol.147 (SMART DSC-2017)

also the classification of written characters, to fault detection in machinery and diagnosing. The act of recognition will be divided into 2 broad categories; those are recognizing concrete things and recognizing abstract things. The popularity of concrete things involves the popularity of spatial and temporal things. Pattern recognition will be done each in traditional computers and neural networks. This technique can provide result as affirmative or NO. That’s why this is often an easy technique. The speedily growing and out there computing power, whereas sanctioning quicker process of big knowledge sets has conjointly expedited the utilization of elaborate and numerous ways for knowledge analysis and classification [4]. At an equivalent time demands on automatic pattern recognition systems are rising staggeringly owing to the provision of enormous databases and rigorous performance needs i.e. speed, accuracy, and cost.

2 Existing Work Neural network classification: The word Neural Network‟ has been motivated from its inception by the recognition that the human brain computes in an entirely different way from the Conventional digital computer. The brain is a highly Complex, non-linear and parallel computer (information processing system). An Artificial Neural Network (ANN), often just called a "Neural Network" (NN), is a mathematical biological neural network [1]. It consists of an interconnected group of artificial network and processes information using a connectionist approach to computation. Image recognition: There are various techniques for currency recognition that involve texture, pattern or color based. We use digital image processing techniques to find region of interest, after that Neural Network and Pattern Recognition Technique is used for matching the pattern. Microcontroller is used to interface the internet access for the time which is programmed[2]. At once the time is about to be over the controller asks for further ng access to the internet actions. The steps are:  Scanning the image at 150 dpi with 128x128 pixels by simple flat scanner.  ROI is extracted  Converting image to gray scale image and setting up a level.  Applying sobel edge filter, average filter, laplacian filter and getting reference value.  After obtaining reference value, it is matched by using Neural Networks and Pattern Recognition Tool in Mat lab, if the image is matched then internet access is provided. Pattern recognition methods: Pattern recognition include a lot of methods which impelling the development of numerous applications in different filed. Models opted for pattern recognition can be categorized in to different categories depending upon the method used for data analysis and classification. Models can be independently or

Copyright © 2017 SERSC

435

Advanced Science and Technology Letters Vol.147 (SMART DSC-2017)

dependently used to perform a pattern recognition task. The different models used for pattern recognition task are given as follows     

Statistical Model Structural Model Neural Network Based Model Fuzzy Based Model Hybrid Model

Statistical model: In Statistical method of Pattern Recognition each pattern is described in terms of features. Features are chosen in such a way that different patterns occupy non-overlapping feature space. It recognizes the probabilistic nature both of the information we seek to process, and of the form in which we should express it. It works well when the selected features lead to feature spaces which cluster in a recognizable manner, i.e. there is proper interclass distance. After analyzing the probability distribution of a pattern belonging to a class, decision boundary is determined. Here patterns are projected to pre-processing operations to make them suitable for training purposes. Features are selected upon analyzing training patterns. Test patterns are applied to check suitability of system to recognize patterns. Feature measurement is done while testing, then these feature values is presented tolerated system and in this way classification is performed.

Fig. 1. Statistical model

Structural model: Structural pattern recognition systems have proven to be effective for data which contains an image data and time series data (which is organized by time). The usefulness of structural pattern recognition systems, however, is limited as a consequence of fundamental complications associated with the implementation of the description and classification tasks.

436

Copyright © 2017 SERSC

Advanced Science and Technology Letters Vol.147 (SMART DSC-2017)

Fig. 2. Structural model

Some structural pattern recognition systems justify the use of a particular set of feature extractors by claiming that the same set had been used successfully by a previous system developed for a similar application within the same domain; such claims simply shift the burden of feature extractor development onto previously implemented systems. Simplistic primitives are domain independent, but capture a minimum of structural information and postpone deeper interpretation until the classification step. Domain and application specific primitives can be developed with the assistance of a domain expert, but obtaining and formalizing knowledge from a domain expert, called knowledge acquisition, is very difficult.

Fuzzy based model: Syntactic techniques are utilized when the pattern sought is related to the formal structure of language. Semantic techniques are used when fuzzy partitions of data sets are to be produced. Then a similarity measure based on weighted distance is used to obtain similarity degree between the fuzzy description of unknown shape and reference shape. Hybrid model: In most of the emerging applications, it is clear that a single model used for classification doesn’t behave efficiently, so multiple methods have to be combined together giving result to hybrid models. Primitive approaches to design a Pattern Recognition system which aims at utilizing a best individual classifier have some drawbacks. It is very difficult to identify a best classifier unless deep prior knowledge is available at hand. Statistical and Structural models can be combined together to solve hybrid problems. In such cases statistical approach is utilized to recognize pattern primitives and syntactic approach is then used for the recognition of sub-patterns and pattern itself. Fu gave the concept of attributed grammars which unifies statistical and structural pattern recognition approach. To enhance system performance one can use a set of individual classifiers and combiner to make the final decision.

Copyright © 2017 SERSC

437

Advanced Science and Technology Letters Vol.147 (SMART DSC-2017)

The multilayer neural network match each pixel of given sample and provide the exact match.

Fig. 3. Matching process

3 The Design of Interface The main element is the controller interface ENC28J60 made by Microchip company. It is designed to serve as an Ethernet network interface for any controller equipped with SPI. ENC28J60 meets all specifications of IEEE 802.3 having incorporated MAC and PHY modules. It also provides an internal mode DMA for fast data transfer and hardware for TCP/IP. The Enc28J60 has a clock at 25MHz. The Microcontroller is programmed in a way that it controls the accessing of internet. The time for access is programmed for each currency. Each currency’s image is already stored in the memory. Based on the given input the time is allotted. For example the 50 rupee note is programmed for accessing 15 minutes.

438

Copyright © 2017 SERSC

Advanced Science and Technology Letters Vol.147 (SMART DSC-2017)

Fig. 4. Pin configuration of ENC28J60

Once the note is inserted the image processing is started and once the currency’s image is matched with the reference image then a signal is sent to the microcontroller. The microcontroller is used to control the time management and accessing internet. The time is programmed and it allocated based on the input. Once the time is about to expire, the microcontroller asks the user whether to continue or not. If the user wishes to continue, another currency has to be inserted. Now again the image processing is started. Once the user doesn’t wish to continue, the Microcontroller stops accessing the internet. Suppose the user has only a 100 rupee note and needs to operate only for 15minutes (for only 50 rupees), then the user has to get the balance 50 rupees. This is also programmed in the Microcontroller to provide the balance money. The corresponding currency will be sent out through the control of Microcontroller. This also includes the image processing technique. The aim to show how to make Ethernet interface with microcontroller automation systems. Are given at all stages of design, implementation and programming interface. The only requirement is that the microcontroller to have SPI interface (Serial Peripheral Interface). SPI is a standard synchronous high speed interface that operates in full duplex and can operate with one master device and one or more slave devices.

4 Conclusion The Microcontroller interface now offers an effective solution, making low-cost (about 10euro), robust in operation. The system can be used for monitoring and remote control is accessible through any Internet connection. We can connect to the microcontrollers or microprocessors and other thereby 2371 improving data processing speed. Ordering system can be used remotely via the internet to different facilities or equipment used for household use. A comparative view of all the models of pattern recognition has been shown which says that for various domains in these areas different models can be used. In case of noisy patterns, choice of statistical model is a good solution. Structural model depends upon recognition of simple pattern primitives and their relationships represented by description language. To recognize unknown shapes fuzzy methods are good options. Hybrid model is

Copyright © 2017 SERSC

439

Advanced Science and Technology Letters Vol.147 (SMART DSC-2017)

combination of both statistical and structure model. So it is best method to solve the many problems. We use hybrid error model for super resolution of images.

References 1.

Amel Saeed Tuam” Iris Image Segmentation and Recognition” Int. J Comp Sci. Emerging Tech Vol-3 No 2 April, 2012. A. Ms.Trupti Pathrabe and B.Dr. N.G.Bawane, “Paper Currency Recognition System Using Characteristics Extraction and Negativity 2. Correlated NN Ensemble, 2010, Int. Journal of Latest Trends in Computing. 3. Statistical pattern recognition: A Review; Robert, Jain, Mao, IEEE transaction on machine learning and pattern analysis, VOL.22. NO.1. Jan 2000. 4. Surjeet Singh, KulbirSingh “Segmentation Techniques for Iris Recognition System,” International Journal of Scientific & Engineering Research Volume 2, Issue 4, April-2011 5. Zheng, He; Classification technique in pattern recognition; faculty of IT, University of technology, January 31 – February 04, 2009. 6. SeemaAsht,Rajeshwar Dass, “Pattern recognition Techniques :A Review, International Journal of Computer Science and Telecommunications- Volume 3, Issue 8, August 2012. 7. Harish Agarwal, Padam Kumar, “Indian Currency Note Denomination Recognition in Color Image, Int. Journal on Advanced Computer Eng. and Communication Tech. Vol.1. 8. Florin RAVIGAN, Niculae BOTEANU, Adrian Mircea DRIGHICIU, “Ethernet interface for a system of automation with microcontroller”2011. 9. A. Murugan “Fragmented Iris Recognition System using BPNN” International Journal of Computer ApplicationsVolume 36– No.4, December 2011 10. D.A.KS.Gunaratna,N.D.Kodikara,H.L.Premaratne, ” ANN Based currency Recognition System using Compressed Gray Scale and Application for Sri Lankan currency NotesSLCRec”, World Academy of Science, Engineering and Tehnology,2008. 11. CAO Bu-Qing1,2,3LIU Jian-X1 un, “Currency Recognition Modeling Research Based on BP Neural Network”,2010.

440

Copyright © 2017 SERSC