Velagapudi RamaKrishna

Siddhartha Engineering College (Autonomous) Department of Information Technology

7053 Mini Projects Abstracts

October, 2010

Mini Project Advisors

A.Koteswara Rao M.Suneetha V.RadheSyam B L N Phanindra K.Sita Kumari S.Suhasini P.MadhaviLatha M.Ashok Kumar T.Lakshmi Surekha Y.Sangeetha G.Geetha S.Sunitha G.JayaLakshmi

Department of Information Technology, 2010

Table Of Contents
S.No 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. Title Automatic Number Plate Recognition Edge Detection Optical Character Recognition Mining Frequent Closed Sequences without Candidate Generation Segmentation and Character Recognition Based on Euler Numbers Gender Detection Lane Detection ContourDetection-Freeman Chain Code Localization of Number Plates Contour Detection Face Recognition Outliers in Statistical Data Piece Extraction Implementation of Partition Based Clustering GHIC:A Hierarchical pattern Based Clustering Algorithm for Grouping Web Transactions Sequential Pattern Mining of Customer Transactions Skew Detection Single Document Sumarization Back Ground Segmentation Contrast Enhancement Sampling Based Algorithm for Discovering Association rules Implementing partitioning approach Clustering Techniques for Huge Dimensional DataBase Eye Detection LAN Detection using OpenCV Implementation of DBScan Algorithm Mining Frequent closed item sets::CHARM Algorithm Text Mining With Information Extraction Web Search Engine by Implementing Ranking and Clustering Techniques Contour Detection Page No 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36

Department of Information Technology, 2010

Table Of Contents
30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. Face Detection Morphological Operations Optical Character Recognition Contrast Enhancement-Histogram Equalization Localization of Number Plates Motion Detection Number Plate Localization using Morphological Process Skew Detection Edge Detection Localization of Number Plates Optical Character Recognition 37 38 39 40 41 42 43 44 45 46 47

Department of Information Technology, 2010

Computer Vision

Automatic Number Plate Recognition

B.LakshmiKanth(Y07IT008), K.Aditya(Y07IT043)

ANPR (Automatic Number Plate Recognition) is the process of detecting and recognizing the license plate of vehicles. The concept behind why such systems are developed is, with the advent of population growth maintenance of security levels are becoming very difcult.The growth of population needs a good road infrastructure. When the infrastructure is good it is obvious that sales in automobiles increase. So to maintain security some automation is required. The eld which provides such automation and decreases the human effort is Intelligent Transport Systems (ITS). One of the outcomes of this eld is ANPR systems. These systems can process a video or an image and gives us the facts. ANPR uses the image processing techniques.

Department of Information Technology, 2010

Computer Vision

Edge Detection
P.Pardhasaradhi(Y08IT137), N.Praveen Kumar(Y07IT87), B.Ramesh(Y07IT96), A. Ashok(Y07IT110)

Edge detection refers to the process of identifying and locating sharp discontinuities in an image. The discontinuities are abrupt changes in pixel intensity which characterize boundaries of objects in a scene. Classical methods of edge detection involve convolving the image with an operator (a 2-D lter), which is constructed to be sensitive to large gradients in the image while returning values of zero in uniform regions. There is an extremely large number of edge detection operators available, each designed to be sensitive to certain types of edges. Variables involved in the selection of an edge detection operator include: Edge structure: Not all edges involve a step change in intensity. Effects such as refraction or poor focus can result in objects with boundaries dened by a gradual change in intensity. The operator needs to be chosen to be responsive to such a gradual change in those cases. Newer wavelet-based techniques actually characterize the nature of the transition for each edge in order to distinguish, for example, edges associated with hair from edges associated with a face.The purpose of detecting sharp changes in image brightness is to capture important events and changes in properties of the world. It can be shown that under rather general assumptions for an image formation model, discontinuities in image brightness are likely to correspond to discontinuities in depth,discontinuities in surface orientation,changes in material properties and variations in scene illumination.The Sobel operator performs a 2-D spatial gradient measurement on an image and so emphasizes regions of high spatial frequency that correspond to edges. Typically it is used to nd the approximate absolute gradient magnitude at each point in an input grayscale image.Detect the edges by loading the image..The actual process is the image can be converted in to matrix.. based on intensity values..This image matrix can be multiplied with sobel kernels.In both horizontally and vertically .and then the two matrix combined ,to form the edges of an image. . . . Outcomes:Edges of an input image is detected. . . . . .

Department of Information Technology, 2010

Computer Vision

Optical Character Recognition

P.S.Sai Lakshmi(Y07IT080) , A.Sushma(Y07IT107) , P.S.Sai Meenakshi(Y07IT081)

Full form of OCR is Optical Character Recognition. This technology allows a machine to automatically recognize characters through an optical mechanism.It is a computer program designed to convert scanned or digital images of handwritten or typewritten text into machine-editable text, or to translate pictures of characters into a standard encoding scheme representing them (e.g. ASCII or Unicode). OCR software can recognize a wide variety of fonts.Optical character recognition (OCR) refers to a computers ability to recognize printed letters, numerals, or symbols (optical characters) as discrete entities rather than as simply an image containing lines, curves, and shading.It is widely used to convert books and documents into electronic les, to computerize a record-keeping system in an ofce, or to publish the text on a website.OCR makes it possible to edit the text, search for a word or phrase, store it morecompactly.

10

Department of Information Technology, 2010

Data Engineering

Mining frequent closed sequences with out candidate generation

K. Vivek Manohar(Y07IT034) , K. Brahma reddy(Y07IT045), K. Tarak Ravi(Y07IT037), M.sai kiran(Y07IT063)

The data is being increased in day-to-day business application. We are storing all the data using the databases. But there is no use of just storing the data in a database. If we analyze the data we can nd some knowledge about the data. So for analyzing of data a new concept data mining is introduced. Using data mining we can nd the relationships between the data variables. In mining we have a number of techniques to nd relationships between data. It is very useful in any business to analyze the data. Business such as supermarket,Bank,Astronomy,Biology etc. can use data mining for analyzing the data.The relationships between the data can be found using the FP-Growth algorithm.What we are going to nd is the frequently occurring patterns.For example let us take the sales in a shopping market.By analyzing the sales the owner can know the item having more demand. And relationship between items such as milk and bread. This application helps us to nd these relationships and make the owner understand the item-item relation.We are a bit confused for nding frequent items as we have a number of techniques such as Association rule mining,Apriori algorithm ,FP-Growth etc.In most of the techniques we have some problems such as time constraint, space constraint etc.But in fp-growth algorithm we avoid them by the help of a tree which is constructed using the recurssion thechnique.So this is the best technique to nd the frequent patterns.Finding frequently occurring events is the big problem in business organizations . If we cannot nd them it leads to failure in the bussiness. In our Supermarket example if the owner doesnot know the items which are sold frequently he may not bring the new stock. This type of problem can be solved easily by applying the FP-Growth algorithm. It simplies the data given by constructing a tree which is easy to understand by all.The major objective of the study is to nd the frequently ouccuring patterns or events and also to nd relationship between the items.By doing so it will be helpful to the bussiness organization and lead them in a way to prot.The outcomes of the fp-growth algorithm is an fp tree that containing the frequent patterns. Using the constructed fp tree the frequent patterns are obtained.

Department of Information Technology, 2010

11

Computer Vision

Segmentation and Character Recognition of License Plates Based on Euler Number

D.Satyavathi(Y07IT020), Ch.Baby Priyanka(Y07IT011), K.Bhaskar(Y07IT015),B.Vishnuvardhan Babu(Y07IT010)

Vision is the process of discovering what is present in the world and where it is by looking. Computer Vision is the study of analysis of pictures and videos in order to achieve results similar to those as by people.Our project Segmentation and character recognition based on Euler number is mainly concerned with how the characters are recognized given a localized input image. This is a concept related to OCR, commonly known as Optical Character Recognition, is the mechanical or electronic translation of scanned images of handwritten, typewritten or printed text into machine-encoded text. This is one of the important steps in the Automatic License Plate Recognition (ALPR) of Intelligent Transportation Systems (ITS).The main objectives of our project include Segmentation, describes how the plate is divided into individual characters. It can be done using Scanning, projection method or connected components labeling. We implemented it using connected components labeling. Connected component labeling works by scanning an image, pixel-by-pixel (from top to bottom and left to right) in order to identify connected pixel regions. This connectivity may either 4 or 8. Feature Extraction, is the process of nding out the specic features of each segmented part based on calculating the properties like area, centroid, bounding box and Euler number. Here the property we considered is Euler number, where Euler number is calculated using the formula, Euler Number = Number Connected components - Number of holes.Recognition, is the process of recognizing characters based on properties of image obtained in the earlier step like getting Euler number etc., This step can be implemented by using techniques like Template matching, Fuzzy Logic and Neural Networks etc., Here we implemented recognition based on Template Matching, where the required template (input) is compared with the Target image (set of all alphabet and numbers) to recognize the character. The nal outcome of our project is the recognized characters.

12

Department of Information Technology, 2010

Computer Vision

Gender Detection
J.Kranthi Pradeepika (Y07IT047),M.Krishna Prasanna (Y07IT048)

Automatic gender detection through facial features has become a critical component in the new domain of computer human observation and computer human interaction (HCI). Many examples which include user-interfaces that can detect their presence and count of the users; video security systems or close circuit cameras can record facial images of individuals or groups after their unauthorized entry; now indexing the taken image or video content can be beneted from meta-data from gender detection by the extraction of facial features. This is an extension to fast face gender detection. Our new updates which includes locate each facial feature and based on that gender determination is done.

Department of Information Technology, 2010

13

Computer Vision

Lane Detection
M.RaShaik(Y07IT058),M.Chaitanya Kumar(Y07IT056),K.Deepak Sai Santosh(Y07IT033),K.Sarath Chandra(Y07IT49)

Car accidents kill about 2, 00,000 people each year in the India. Up to 90percent of these accidents are caused by driver faults. Automating driving may help reduce this huge number of human fatalities. One useful technology is LANE DETECTION, which has received considerable attention since the mid 1980s. Techniques used varied from using monocular to stereo vision, using low level morphological operations to using probabilistic grouping and B-snakes.Due to day by day increase in population, there are number of vehicles increasing on the roads. As a result, number of accidents is growing day by day. So the Lane detection presents an effective solution to tackle this problem.Many researchers have shown lane detectors based on a wide variety of techniques. Techniques used varied from using monocular to stereo vision using low level morphological operations to using probabilistic grouping and B-snakes .Finally all these techniques are classied under two main headings.feature based techniques, model based techniques. Our project uses feature based technique and some areas of model based technique. The main aim of our project is to perform lane detection in Indian context, which is really difcult when compared to other countries where this technique is in use. We adopted a stepwise process that has namely reading input image, thresholding, clearing noise and unnecessary boundary objects, identifying the remaining regions and nding the lanes among them. Our project is starting step for a complex process and the code is executed under several circumstances in which above 90percent cases have a threshold value greater than 210. The process effectively applies to highways and the roads that have well painted roads.

14

Department of Information Technology, 2010

Computer Vision

Contour Detection- Freeman Chain Code

T.Lahari(Y07IT051),B.Meghana(Y07IT062),M.Sneha(Y07IT060)

Computer vision is a scientic discipline, concerned with the theory behind articial systems that extracts information from images. The elds closely related to computer vision are image processing, image analysis and machine vision. The term Contour can be dened as an outline or a boundary of an object.Hence, Contour detection deals with detecting various objects in an image specically. Use of contour detection in image processing is to locate objects and their boundaries in images.Recognition of objects is one of the basic tasks in computer vision applications. Recognition is usually based on gray levels or colours, and shape characteristics of target objects. The goal of contour recognition is to nd a description of objects specically so as to distinguish between different target objects. The importance of contour detection is it detects the salient boundaries of the object in an image.

Department of Information Technology, 2010

15

Computer Vision

Localization of Number Plate

Anusha.K (Y07IT006),Mownica.K (Y07IT036),Sravani.K (Y07IT041)

Number plate detection and recognition is an image-processing technique used to identify a vehicle by its number plate. Technological intelligence is a highly sought after commodity even in trafc-based systems. These intelligent systems do not only help in trafc monitoring but also in commuter safety, law enforcement and commercial applications. In this paper, a number plate localization system for vehicles is proposed. This system is developed based on digital images and can be easily applied to commercial car park systems for the use of documenting access of parking services, secure usage of parking houses and also to prevent car theft issues. There are several algorithms that can be used in order to detect the number plate in various formats such as Morphology, Thresholding (Adaptive thresholding) , Hough transforms , Edge detection and Filtration techniques etc. The proposed license plate localization algorithm is based on a combination of Thresholds with a Hough Rectangular Transform approach.

16

Department of Information Technology, 2010

Computer Vision

Contour Detection
N.Vineela Kumari(Y07IT125),P.Pavan Kumar(Y07IT082),V.Sri Vigna Harish(Y07IT120),E V Venkateswara Rao(Y08IT140)

Contour Detection has been a eld of interest in recent years. The contour detection is nothing but the detection of boundary for a given image. It is the basic operation for many image processing techniques such as Eye Detection, Image Segmentation, Lane Detection, Motion Detection, Localization of a given image, Object recognition and recognition. As a result the detection of contour has become a pre- requisite in most of the image processing techniques. Initially edge detection is used in order to nd the boundaries of a given image. In edge detection any change in the intensity value of the image is marked as an edge. So the conventional edge detectors may detect many spurious edges in textured regions where there is no boundary. For example, in the case of many x-rays where the image is an ultrasound textured image, it is not easy to differentiate the parts present in it. So it denitely demands a boundary between them to make them obvious.Many approaches are present in order to nd the contour of a given image. Some of the approaches are the Bayesian Contour detection, Texture- based Contour Detection. Many algorithms were proposed for the detection of contour in a image such as the Snakes algorithm, Hough Transform algorithm, Geometric Active Contours (GAC), Level Set Method Approach etc. The efciency and accuracy depends on the point of application of the method and the method used to nd the contour in that context. The main objective of the project is to nd the contour or boundary of a given image efciently, effectively and accurately. In many applications it is difcult to nd the image contour due to the variations in the intensity levels of the image.The greater the intensity of the image the easier to nd the contour of the image. Our objective is to nd the contour even for an image of lower intensity value. It is difcult to nd the contours of curved and elliptical surfaces. Our project strives for nding the contours of the curved and elliptical surfaces as well.This project is used to nd the contours for an image that can be captured by the webcam or camera. Mainly it is used in the security purposes for authentication services. This project works, based on the threshold value and the intensity of the picture that has captured by the camera or webcam. If the intensity of the captured image is good then we are able to nd the contours effectively. If the intensity of the captured image is not good ie. in the night vision it is not efcient to nd the contours.

Department of Information Technology, 2010

17

Computer Vision

Face Recognition
Soumya.Siripurapu (Y06IT052),G.Alekhya (Y07IT027),P.Lidia (Y06IT025),Ch.Ramesh SaiVarma (Y05IT034)

This paper describes the design and construction of a test-bed for prototyping of embedded face detection and recognition algorithms. The test-bed is called the PITS (Portable Interactive Terrorist Identication System), it makes use of a hand-held device called the Sharp Zaurus. The embedded device has a processor, camera, color display, and wireless networking. This system is different from existing systems because of its embedded information technologies. The embedded device performs both detection and recognition. We present a skin color approach in the YCbCr color space for fast and accurate skin detection. We then process this image using a combination of morphological operators and elliptical shape of faces to segment faces from the other skin colored regions in an image. An eigenface algorithm processes the segmented faces and matches the face to a face database.

18

Department of Information Technology, 2010

Data Engineering

Outliers in Statistical Data

Ch.Ramesh Babu(Y07IT017),Rahaman(Y07IT065),G.Srikanth Reddy(Y07IT025),Pradeep(Y07IT064)

More data is generated by the Bank, Business, astronomy, biology etc. In that data there is some useful data and un useful data.Outlier mining is one of important research elds of data mining, used to nding a small part of data objects in data set which significantly differ from other objects in common behaviors and data mode.Finding the credit card fraud had become a major problem in every bank organization. This leads to serious problem in the organization.Here we treat that fraud as a outlier and nd that using the distance based algorithm.Some of the algorithms to nding the outliers are: Statistical distribution based outlier detection.The statistical distance based approach to outlier detection assumes distribution or probability model for the given data set and the identies outliers with respect to model using a discordances test.According to the density based approach an object is a local outlier if it is outlying Relative to it local neighborhood particularly with respect to density of the neighborhood.Deviation based method identies outliers by examining the main characteristics of the objects in a group. Object that from these description are consider outliers.The project mainly deals with the nding the Credit card fraud. This is very useful in the banking sector. The output of the distance based algorithm is the simple which gives customers who are involving in the credit card fraud. It display the fraud customers.

Department of Information Technology, 2010

19

Computer Vision

Piece Extraction
V.Anish(Y07IT005),K.Lakshmi RaviKiran(Y07IT053),M.V.SaiKumar(Y07IT061),P.Gowtham(L08IT134)

The next step after the detection of the number plate area is a segmentation of the plate. The segmentation is one of the most important processes in the automatic number plate recognition, because all further steps rely on it. If the segmentation fails, a character can be improperly divided into two pieces, or two characters can be improperly merged together.We can use a horizontal projection of a number plate for the segmentation, or one of the more sophisticated methods, such as segmentation using the neural networks. If we assume only one-row plates, the segmentation is a process of nding horizontal boundaries between characters. Section 3.2 deals with this problematic. The second phase of the segmentation is an enhancement of segments. The segment of a plate contains besides the character also undesirable elements such as dots and stretches as well as redundant space on the sides of character. There is a need to eliminate these elements and extract only the character. Section 3.3 deals with these problems.

20

Department of Information Technology, 2010

Data Engineering

Implementation of Partition-Based Clustering

K.Kavya(Y07IT039),K.Pavani(Y07IT035),K.Rajitha(Y07IT042), E.Thripuana Kumari(L08IT132)

Day by day the size of data thats being stored is increasing .In order to obtain knowledge from massive data we need a mechanism, Data mining is such mechanism which extract patterns (knowledge) from data.( Data could be large in two senses: in terms of size or in terms of dimensionality.).There are various data mining techniques among them most widely used technique is clustering. Clustering is used to place data elements into related groups without advance knowledge of the group denitions.Our project is to implement partition-based clustering. The partition-based clustering generally result in a set of M clusters, each object belonging to one cluster. Among these partition-based clustering techniques K-means technique is simple to implement and easy to debug. So we initially implemented this technique on a database containing multiple choice questions and designed an application called computer adaptability test to illustrate how the partitioned data can be used in real life.

Department of Information Technology, 2010

21

Data Engineering

GHIC: A Hierarchical Pattern-Based Clustering Algorithm for Grouping Web Transactions

A.Preethi Latha(Y07IT088),Y.Priyanka(Y07IT089), B.Vasavi(Y07IT119)

Data mining, is a process of extracting useful data from large database, it is a powerful technology with great potential to help companies focus on the most important information in their data warehouses. Data mining is seen as an increasingly important tool by modern business to transform data into an informational advantage. Data mining tools predict future trends and behaviors, allowing businesses to make proactive, knowledgedriven decisions.Web mining can be broadly dened as the discovery and analysis of useful information from the World Wide Web. There are roughly three knowledge discovery domains that pertain to web mining: Web Content Mining, Web Structure Mining, and Web Usage Mining. Web usage mining is also known as Web Log Mining, is the process of extracting interesting patterns in web access logs. Clustering: The process of dividing a dataset into mutually exclusive groups such that the members of each group are as close as possible to one another, and different groups are as far as possible from one another, where distance is measured with respect to all available variables. Applications of Web Mining:E-commerce,SearchEngines,Website Design

22

Department of Information Technology, 2010

Data Engineering

SPM: Sequential pattern mining with candidate generation Implementation using JAVA
D.Srujana(Y07IT021),M.Sowmya(Y07IT057),U.Jyothi(Y07IT032)

Datamining is the extraction of hidden predictive information from large databases, is a powerful new technology with great potential to help companies focus on the most important information in their data warehouses. Data mining tools can answer business questions that traditionally were too time consuming to resolve. Apriori is the most famous and basic method in mining association rules. As time passes by, new datasets are inserted, old datasets are deleted while some other datasets are updated. It is obvious that time stamp is an important attribute of each dataset, also it is important in the process of data mining and it can give us more accurate and useful information. Sequential pattern mining is trying to nd the relationships between occurrences of sequential events, to and if there exist any specic order of the occurrences. Sequential pattern is a sequence of item sets that frequently occurred in a specic order, all items in the same item sets are supposed to have the same transaction-time value or within a time gap. Usually all the transactions of a customer are together viewed as a sequence, usually called customersequence, where each transaction is represented as an item sets in that sequence, all the transactions are list in a certain order with regard to the transaction-time.

Department of Information Technology, 2010

23

Computer Vision

Skew Detection
G.Harshavardhan(Y07IT030),C.Harika(Y07IT029),G.Prasanna Kumar(Y07IT026),K.Nanchara Rao(Y07IT046)

When an image or a document is loaded or scanned there may be some changes from the original image. These changes may be in color, size, skew etc. Skew is an artifact that can occur in scanned images because of the camera being misaligned, imperfections in the scanning or surface, or simply because the paper was not placed completely at when scanned.Skew is the deviation in angle of a captured image. It signicantly degrades the recognition abilities of the image. In this project we aim at detecting skew in an image.Skew detection is one the rst operations to be applied to scanned documents when converting data to a digital format. Its aim is to align an image before processing because text segmentation and recognition methods require properly aligned next lines. We chose Hough Transform in order to detect skew. The classical Hough transform is most commonly used for the detection of regular curves such as lines, circles, ellipses, etc. The Hough transform can also be used to compute an approximate angle of image. The Hough Transform transforms the Cartesian space X-Y into the Hough Domain ( , ). It is performed on the scanned image and the variance in ? values is calculated for each value of ?. The angle that gives the maximum variance is the skew angle.This skew detection and correction of the image by the skew angle will achieve a high character recognition accuracy rate, as highest performance is achieved from skew-free scanning. Skew detection has many applications in the Automatic Number Plate Recognition [ANPR], Face recognition, Document- Correction, Optical Character Recognition [OCR], Optical Mark Recognition, barcode detection etc.

24

Department of Information Technology, 2010

Data Engineering

Document Summarization
Y.Samuel Johnson(Y07IT131),Vinay Badri(Y07IT130),Pavan Kumar(Y07IT114),Phani Kumar(Y07IT115)

A document-summarization system takes a document and automatically highlights a small set of sentences that are expected to cover the different aspects of the document. The sentences are picked using simple coverage and orthogonality criteria. It gives overview of the document without having to spend lots of time drilling down into the details. The system handles plain text. It runs as a stand-alone application. For comparative analysis and exibility, the system includes other text-summarization methods, e.g. k-median clustering and singular value decomposition etc. Thus, the system allows the user to explore the content of the input document in many different ways.

Department of Information Technology, 2010

25

Computer Vision

BackGround Segmentation
G.MuraliKrishna(Y07IT070),K.MohanChand(Y07IT066), B.Prasanth(Y07IT084),R.Rajesh(Y07IT095))

We present a real-time algorithm for background segmentation[1]. Sample background values at each pixel are quantized into codebooks which represent a compressed form of background model for a long image sequence. This allows us to capture structural background variation due to periodic-like motion over a long period of time under limited memory. The codebook representation is efcient in memory and speed compared with other background modeling techniques. Our method can handle scenes containing moving backgrounds or illumination variations, and it achieves robust detection for different types of videos. Object detection and tracking has various application areas including intelligent transportation systems. These object detection and tracking approach that combines the background subtraction algorithm and the feature tracking. There is no project running in openCV without segmentation

26

Department of Information Technology, 2010

Computer Vision

Contrast Enhancement
P.Jithin(Y07IT078),V. Seshu (Y07IT100),K. Vaishnav (Y07IT116), P. Vineetha (Y07IT126) ))

Computer vision is concerned with the theory behind articial systems that extract information from images. A branch of articial intelligence and image processing Concerned with computer processing of images from the real world. Computer vision typically refers to enhance the image quality (e.g. remove noise, increase contrast) and image understanding to recognize features present in the image. It is the science and technology of machines that see, where see in this case means that the machine is able to extract information from an image that is necessary to solve some task. The image data can take many forms, such as video sequences, views from multiple cameras, or multi-dimensional data from a medical scanner. As a technological discipline, computer vision seeks to apply its theories and models to the construction of computer vision systems. The computer vision enables the images that acquired from the real world to represent the objects. The goal of Computer vision is to process images acquired with cameras in order to produce a representation of objects in the world. Computer vision gives the idea of the information in the images that cannot be viewed by the naked eye. There may be some hidden information in the images which we cannot see this can be known with the help of computer vision. Examples of applications of computer vision include systems for: Controlling processes (e.g., an industrial robot or an autonomous vehicle). Detecting events (e.g., for visual surveillance or people counting). Organizing information (e.g., for indexing databases of images and image sequences). Modeling objects or environments (e.g., industrial inspection, medical image analysis or topographical modeling). Interaction (e.g., as the input to a device for computer-human interaction).

Department of Information Technology, 2010

27

Computer Vision

Association Rule Mining

K.V.KrishnaRao(Y07IT040),D.Vinay Prakash(Y07IT022),E.Sravan Sandeep(Y07IT023),K.Naveen Kumar(Y07IT050) )

Association Rule mining is one of the classical DM technique. Association Rule mining is a very powerful technique of analysing / ndingpatterns in the data set. Association rules show attribute value conditions that occur frequently together in a given dataset. The classic application of the association rule mining is to analyse the Market Basket Data of a retail store. It is mainly used to nd relation used between two items. Association rules provide information of this type in the form of if-then statements. Sampling is a powerful data reduction technique that has been applied to a variety of problems in database systems.

28

Department of Information Technology, 2010

Data Engineering

Implementing Clustering Techniques

P.Mahesh(Y07IT059),P.Devi Sai Jyothi(Y07IT018), Ch.Vijay Kumar(Y07IT016) ) )

More data is generated by the Bank, Business, astronomy, biology etc. Everyone is need to ne familiar with the Decision Making Process. We all rely on information, and techniques or tools, to help us in our daily lives. When we go out to eat, the restaurant menu is the tool that provides us with the information needed to decide what to purchase and how much to spend. Operating a business also requires making decisions using information and techniques - how much inventory to maintain, what price to sell it at, what credit arrangements to offer, how many people to hire. Decision Making Process in business is the systematic process of identifying and solving problems, of asking questions and nding answers. Decisions usually are made under conditions of uncertainty. The future is not known and sometimes even the past is suspect. This guide opens the door for business owners and managers to learn about the variety of techniques which can be used to improve your decision making process in a world of uncertainty, change, and uncontrollable circumstances.

Department of Information Technology, 2010

29

Computer Vision

Eye Detection-Grey Intensity Face Without Spectacles

N.Greeshma(Y0IT077),B.Tejaswi(Y07IT109), U.Nandini(Y07IT113),N.Vineela(Y07IT125) ) )

The principle objective of this paper is to demonstrate the capability of eye detection in grey intensity images to conduct mapping accurately.Eye detection serves to verify the correctness of face on an image. To make an accurate mapping we make use of feature based and template based techniques. Feature based is used to locate the rough region of face and eyes. Eye detection is dened as to locate the eye position on the given image. From face we can detect the region of eyes. Human eyes play an important role in face recognition and face expression where the eye detection method helps to detect eyes before the detection of other facial expressions. It includes cognitive science, medical research, human interaction, computer usability and so on.

30

Department of Information Technology, 2010

Computer Vision

Lane Detection using OpenCV

D.Sai Chakradhar(Y07IT098),Ch.Prudhvi Krishna(Y07IT090),B.Sri Phani Krishna(Y07IT104),Ismail(L07IT139)) ) )

Autonomous Guided Vehicles (AGV) have found many applications in many industries. Their utilization had been explored in areas such as hospitals for transportation of patients, automated warehouses and other hazardous related areas. In most applications, these AGVs have to navigate in unstructured environments. Path ndings and navigational control under this situation are accomplished from the images feedback by camera mounted on the vehicles. These images are interpreted to extract meaningful information such as positions, road markings,road boundaries, and direction of vehicles heading. Among many extraction methods, lane marking or road boundary detection from these images had received great interest. This is one of the most fundamental approach in detecting the curvature of the following the road. As the images received are usually corrupted by noises, lots of algorithms for boundary detection were developed to achieve robustness in rejection these noises. The main properties that must be possessed by a solution of the lane marking or boundary detection are: The detection should not be affected under shadow condition. These shadows can be cast by trees, buildings, etc. It should be capable of processing painted or unpainted roads. It has to detect painted lines and road boundaries for painted and unpainted road respectively. It should handle curved road rather than assuming straight road. It should use the parallel constraint as a way to improve the detection of both sides of lane markings or boundaries in the face of noise in the images. It should produce an explicit measure of the reliability of the result it has produced.

Department of Information Technology, 2010

31

Data Engineering

DBScan Algorithm
Akesh(Y07IT019),BalaKrishna(Y07IT012),BalaRam(Y07IT009)) ) )

Data mining is the process of extracting patterns from data. Data mining is seen as an increasingly important tool by modern business to transform data into an informational advantage. Data mining, the extraction of hidden predictive information from large databases, is a powerful new technology with great potential to help companies focus on the most important information in their data warehouses.It is the descriptive tasks that groups according to the logical relationships. The groping can be done based on the functionalities the goods or objects exhibit.DBSCAN (Density-Based Spatial Clustering of Applications with Noise) is a data clustering algorithm. It is a density-based clustering algorithm because it nds a number of clusters starting from the estimated density distribution of corresponding nodes. DBScan is one of the most common clustering algorithms and also most cited in scientic literature.

32

Department of Information Technology, 2010

Data Engineering

Mining Closed Frequent Itemsets Using CHARM Algorithm

K.Swarupa(Y07IT108),K.ValliMadhuri(Y07IT117),Y.Vasanthi (Y07IT118)) ) )

Mining frequent patterns or itemsets is a fundamental and essential problem in many data mining applications. These applications include the discovery of association rules, strong rules, correlations, sequential rules, episodes, multi-dimensional patterns, and many other important discovery tasks.This algorithm is mainly used to solve the problems like nding frequent itemsets in large Market basket data and nding similar item in a large collection. For ex nding documents on web that share a signicant amount of common text, to nd the books that have been bought by many of the same Amazon customers.There are various algorithms to nd the closed frequent itemsets in a large collection . Few of them are Close, A-close, Closet, Charm algorithm and we have preferred to implement charm algorithm.The main objective of our project is to nd closed frequent itemsets from large collection of items .This can be effectively done by rst nding the frequent itemsets .Then nd the closed frequent itemsets using the frequent itemsets that were previously found. The best way to implement is by using Charm algorithm.The outcome of this project is nding Closed Frequent itemsets for a given large collection of itemsets.

Department of Information Technology, 2010

33

Data Engineering

Text Mining With Information Extraction

R.SudhaRani(Y07IT092),M.Anusha(Y07IT073),P.Sudheer(Y07IT105)

Text mining is generally referred as text DATA MINING and roughly equivalent to text analysis refers to the process deriving high-quality information from text. The purpose of text mining is to process the unstructured (textual) information, extract meaningful numeric indices from the text, and thus make the information contained in the text accessible to the various data mining algorithms such as statistical and machine learning algorithms. Text mining is the procedure of synthesizing the information by analyzing the relations, the patterns, and the rules among textual data-semi-structured or unstructured text. Information can be extracted to derive summaries for the words contained in the documents or to compute summaries for the documents based on the words contained in them.

34

Department of Information Technology, 2010

Data Engineering

Web Search Engine by implimenting Ranking and Clustering Techniques

CH.SriHarsha(Y07IT103),SK.Mohd Zaheer Parvez(Y07IT101),T Abhishikth(Y07IT111),S Balaji(L08IT138)

It is a common experience to the web users with the existing search engines like Google, Yahoo, MSN, Ask, e.t.c., that return a long ranked list of related results to the given query. Web users have to go through the list and examine the titles and (short) snippets sequentially to identify their required relevant results. This becomes a time consuming task when multiple sub-topics of the given query are mixed together. Organizing Web search results into clusters facilitates users quick browsing through search results. Clustering of search results is a special technique using which we can organize the retrieved results into meaningful groups enlightening the user work. This work deals with the clustering of web results using Sufx tree based approach. The problem we then faced is to how to name the cluster thus formed. We consider mostly repeated phrase in the document tags as cluster name. Thus in short, a web search results that are fetched from the prevailing web search engines and clustering them into some meaningful subtopics is the theme of the Project.

Department of Information Technology, 2010

35

Computer Vision

Contour Detection
G. Prasanth(Y07IT085),V. Hema jyothi(Y07IT122),K. Navya (Y07IT123),D. Rakesh (Y07IT124)

Computer vision is the science and technology of machines that see. It is concerned with the theory behind articial systems that extract information from images. Computer vision seeks to apply its theories and models to the construction of computer vision systems. Examples of applications of computer vision include systems for:Controlling processes (e.g., an autonomous vehicle), Detecting events (e.g., for visual surveillance), Organizing information (e.g, for indexing databases of images and image sequences),Modeling objects or environments (e.g., medical image analysis),Interaction (e.g., input to a device for computer-human interaction).The topic Contour detection comes under Modeling objects or environments.The term Contour can be dened as an outline or a boundary of an object. Hence, Contour detection deals with detecting various objects in an image specically. Use of contour detection in image processing is to locate objects and their boundaries in images.

36

Department of Information Technology, 2010

Computer Vision

Face Detection
V.Venkateswarlu(Y07IT129), N.Gangadhar(Y05IT025)

A real-time face recognition system can be implemented on an IBM compatible personal computer with a video camera, image digitizer, and custom VLSI image correlator chip. With a single frontal facial image under semi-controlled lighting conditions, the system performs (i) image preprocessing and template extraction, (ii) template correlation with a database of 173 images, and (iii) postprocessing of correlation results to identify the user. System performance issues including image preprocessing, face recognition algorithm, software development, and VLSI hardware implementation are addressed. In particular, the parallel, fully pipelined VLSI image correlator is able to perform 340 Mop/second and achieve a speed up of 20 over optimized assembly code on a 80486/66DX2. The complete system is able to identify a user from a database of 173 images of 34 persons in approximately 2 to 3 seconds. While the recognition performance of the system is difcult to quantify simply, the system achieves a very conservative 88

Department of Information Technology, 2010

37

Computer Vision

Morphological Operations
Y07IT044,Y07IT013,Y07IT055

We have studied morphology in Biology. Well, that was biological morphology. This is mathematical morphology. This was one tool that always knew about, but never really used it. For some reason, this was only for some funky effects. These operations that about to explain are super useful!.Mathematical morphology is a technique used not only on images, but also on graphs, meshes, and anything else that has a spatial nature to it.We have two things when working with morphology: The image,The structuring element. The structuring element is much smaller than the image. This structuring element slides over the image and transforms it. The structuring element creates new pixelsor erases them. Based on this, two basic operations are dened: dilation and erosion.

38

Department of Information Technology, 2010

Computer Vision

Optical Character Recognition

T.Nitin(Y07IT112), A.Mukesh (Y07IT069),M.Sravani (Y07IT071)

Computer vision is the science and technology of machines that see, where see in this case means that the machine is able to extract information from an image that is necessary to solve some task. As a scientic discipline, computer vision is concerned with the theory behind articial systems that extract information from images. The image data can take many forms, such as video sequences, views from multiple cameras, or multi-dimensional data from a medical scanner. The goal of Computer vision is to process images acquired with cameras in order to produce a representation of objects in the world.

Department of Information Technology, 2010

39

Computer Vision

Contrast Enhancement-Histogram Equalization

Y07IT004,Y07IT001,Y07IT014 ))

Computer vision is the science and technology of machines that see. It is concerned with the theory behind articial systems that extract information from images. Contrast enhancement is a part of computer vision which models the objects in the environment.Contrast is the difference in the brightness and colour of the object and other objects within the same eld of view. Contrast enhancement increases the total contrast of an image by making light colours lighter and dark colours darker at the same time. We achieve this contrast enhancement by a technique called histogram equalization.The images with low contrast are difcult to study. The details of the image are not clear to extract any information for further processing. Hence histogram equalization It accentuates the details of the image which transforms it into a format that is useful for processing.

40

Department of Information Technology, 2010

Data Engineering

Localization of Number Plates

Y.Prashanthi(Y07IT086),K.Saisree(Y07IT099), M.Sindhuri(Y07IT102)

This paper presents a methodology for extraction of the vehicle number plates from the vehicle images using hybrid mathematical morphology techniques. The main idea is to use different morphological operations in such a way so that the number plate of the vehicle can be identied precisely.The methodology makes the task of extraction of the number plate independent of color, size and location of number plate. The proposed approach involves ve different processes, these are, image enhancement, morphing transformation, morphological gradient, combination of resultant images and extracting the number plate from the objects that are left in the image. This algorithm can quickly and correctly detect the number plate area from the vehicle image. In this study, we introduce a novel method for license plate recognition from a grayscale image of the car. We rst use a preprocessing stage that contains a certain number of transforms established heuristically for reduces the extra information of car picture. Input of the car plate detection and recognition system is the binary image that obtains form reprocessing stage. Next, start a process for identifying and isolation of the cars license plate from the image, then the result of this process send to segmentation part. Segmentation part separates the characters individually. Finally recognition part recognizes the characters giving the result as the plate number.

Department of Information Technology, 2010

41

Computer Vision

Motion Detection
V.SaiKiran(Y07IT121),Ch.SriHarsha(T07IT103), Ch.Mukesh(Y07IT068), G. KrishnaReddy(L08IT141)

This paper presents a motion detection system that uses a mechanism so that the supervisor of a security monitoring system does not have attention to it constantly. In current days, when a security camera is installed, an human being has to be alert all the time while the camera is on since any little movement can require attention of the supervisor. Perception of change in the outside world is of the highest importance for all living beings. In the sense of vision, the great majority of changes in sensory excitation, i.e., in the image, result from motion-of the light source, of the imaged objects or of the observer. Images also change when a light source is turned on or off though this change cant be interpreted as motion.

42

Department of Information Technology, 2010

Computer Vision

Number Plate Localization using Morphological process

A.Harish(Y07IT003),Afroz Sultana(Y07IT002), N.Avinash(Y07IT007)

License plate detection and recognition is an image-processing technique used to identify a vehicle by its license plate. License plate recognition (LPR) is one of the most important applications of applying computer techniques towards intelligent transportation systems (ITS). Trafc monitoring system has now become an essential administrative part in most of the developed and developing countries. In general, such systems monitor/identify the vehicles exceeding speed limits, or monitor the vehicles crossing the stop line at red trafc signal. It may also be used for registering the vehicles getting entry in a shopping mall or in a railway station or in an airport. The key modules of these monitoring systems are: (i) localization of license plates within the image and (ii) recognizing the license number using an OCR system.Since number plate standards are not strictly practiced in India, a large amount of variations are obtained in parameters like, size of number plate and characters, location of number plate, type of font used, background and foreground color, etc. which makes the task of number plate localization all the more difcult. Algorithms used for detecting number plate are: Localization on the Basis of Edge Finding. Localization on the Basis of Window Filtering. Adaptive Thresholding. Morphological Process.(doubt) Hough Transform.

Department of Information Technology, 2010

43

Computer Vision

Skew Detection
B.Navya(Y07IT076),K.Virajitha(Y07IT127)

When an image or a document is loaded or scanned there may be some changes from the original image. These changes may be in color, size, skew etc. Skew is an artifact that can occur in scanned images because of the camera being misaligned, imperfections in the scanning or surface, or simply because the paper was not placed completely at when scanned. Skew is the deviation in angle of a captured image. It signicantly degrades the recognition abilities of the image. In this project we aim at detecting skew in an image.Skew detection is one the rst operations to be applied to scanned documents when converting data to a digital format. Its aim is to align an image before processing because text segmentation and recognition methods require properly aligned next lines. We chose Hough Transform in order to detect skew. The classical Hough transform is most commonly used for the detection of regular curves such as lines, circles, ellipses, etc. The Hough transform can also be used to compute an approximate angle of image. The Hough Transform transforms the Cartesian space X-Y into the Hough Domain ( , ). It is performed on the scanned image and the variance in ? values is calculated for each value of ?. The angle that gives the maximum variance is the skew angle. This skew detection and correction of the image by the skew angle will achieve a high character recognition accuracy rate, as highest performance is achieved from skew-free scanning. Skew detection has many applications in the Automatic Number Plate Recognition [ANPR], Face recognition, Document- Correction, Optical Character Recognition [OCR], Optical Mark Recognition, barcode detection etc.

44

Department of Information Technology, 2010

Computer Vision

Edge Detection
Ch.Mounika(Y07IT067),Mutahar Sultana(Y07IT072), P.Ramesh(y07it091)

The principle objective of this paper is to demonstrate the edge detection. Edge detection is a fundamental tool in image processing and computer vision, particularly in the areas of feature detection and feature extraction. Edges characterize object boundaries and are therefore useful for segmentation, registration, and identication of objects in a scene. Here we detect the edges in many ways by using different algorithms. Here we represent the canny edge detection algorithm. Edge detection refers to the process of identifying and locating sharp discontinuities in an image. The discontinuities are abrupt changes in pixel intensity which characterize boundaries of objects in a scene.There are three variables involved in the selection of an edge detection operator.First one is Edge orientation in which the geometry of the operator determines a characteristic direction in which it is most sensitive to edges.Next is Noise environment.In this both the noise and the edges contain high-frequency content,so Edge detection is difcult in noisy images and the last is Edge structure in which the effects such as refraction or poor focus can result in objects with boundaries dened by a gradual change in intensity.

Department of Information Technology, 2010

45

Computer Vision

Localization of Number Plates

N.SaiRam(Y07IT074), P.KiranKumar(Y07IT079), A.RajeshKumar(Y07IT094), P.SitaRamPrakash(Y07IT083)

We need to have a system that can effectively retrieve. . . or capture the number of the vehicle and retrieve the number plate info at our convenience. We call it Intelligent Transportation System (ITS). Now in our project...we concentrate on locating the accurate position of the number plate of the vehicle once we had a snap of it from the camera.Automatic Number Plate Recognition (ANPR) is a mass surveillance system that captures the image of vehicles and recognizes their license number. ANPR can be assisted in the detection of stolen vehicles. The detection of stolen vehicles can be done in an efcient manner by using the ANPR systems located in the highways. This paper presents a recognition method in which the vehicle plate image is obtained by the digital CCD cameras and the image is processed to get the number plate information.

46

Department of Information Technology, 2010

Computer Vision

Optical Character Recognition

K.Deepak Sai Santosh(Y07IT033), G.Nithesh Raghavendra(Y07IT028),M.LakshmanKumar(Y07IT052),D.EshwarRao(Y07IT024)

Computer vision has been an active area of research for more than three decades. A quick review of the eld reveals that image processing and pattern recognition has been tremendously successful in terms of delivering operational systems. Character recognition techniques associate a symbolic identity with the image of character. Character recognition is commonly referred to as optical character recognition (OCR), as it deals with the recognition of optically processed characters.In a typical OCR systems input characters are digitized by an optical scanner. Each character is then located and segmented, and the resulting character image is fed into a preprocessor for noise reduction and normalization. Certain characteristics are the extracted from the character for classication. The feature extraction is critical and many different techniques exist, each having its strengths and weaknesses. After classication the identied characters are grouped to reconstruct the original symbol strings, and context may then be applied to detect and correct errors The main areas where OCR has been of importance are image processing, automatic number plate recognition (ANPR), text entry (ofce automation), data entry (banking environment) and process automation (mail sorting).Our project starts with scanning a number plate and obtain a scanned image,then the image undergoes pre-processing and segmentation which generates different segmented parts of the image. Each segmented part undergoes feature extraction. In our project we implemented Hus 7 moment invariants method and chain code method of feature extraction. After this stage we obtain the chain codes and hus moments for each segmented part,with the data we proceed for the next step i.e. recognition of the characters

Department of Information Technology, 2010

47