FACIAL EXPRESSION DETECTION AND RECOGNITION
PROJECT GROUP: GORAJ HUMRASKAR, ZAHED KHAN, AMEYA MESTRY, LIJO JOY
GUIDE: PROF.N.A. BHAGAT
Abstract: Automated recognition of facial expression is an important addition to computer vision research because of its relevance to the study of psychological phenomena and the development of human-computer interaction (HCI). We focus on the following:
- Facial Feature Detection: Detecting the facial features such as eyes, eyebrows and mouth.
- Feature Distances and Template Matching: Calculation of various parameters that help in classification.
- Recognition: Identify the facial expression in an image using the results of various computed distances or template matching algorithm.
For the first stage we make use of techniques such as geometrical segmentation of face images and then applying thresholding to extract the features. In the thresholded image we calculate the distance between the eyebrows, the distance between eyes and eyebrows, the opening of the mouth, a total of seven distances. These are compared with a set of average distances calculated for each expression in the database. The set that it matches with is the set it belongs to. In another approach we make use of an average template of the six different expressions and use template matching algorithm to classify expressions in the input image.
 SAMPLING OF OBJECTS THROUGH IMAGE PROCESSING
PROJECT GROUP: MS.SONAL TELI, MS. PEMA CHOZOM THONGON, MR. KAUSHAL AJUGIA
GUIDE: PROF.R.G. SUTAR
Abstract: The aim of this project is to build a robotic arm that will be able to pick up the object of correct colour placed in the conveyor belt. Object will be placed on the conveyor belt, then the camera mounted exactly above the position of the object will take the image of the object, then the image will be processed in the computer using the program written in MATLAB. If the colour of the object matches the intensity of the required colour then the conveyor belt will move forward and drop the object at the end of the belt, but if the object does not match with the required colour then the conveyor belt will move forward and stop near the robotic arm then the robotic arm will pick it up and place it aside. Like this faulty objects are taken aside. We have chosen MATLAB since it has direct functions for image analysis. Also we have used ATMEL89C2051 along with all other hardware to control the movement of the robotic arm. The project involves simulation of the 89C2051 program in Kiel, interfacing of microcontroller with PC and operation of all the hardware components. The report details the results observed during the running the program and proper functioning of the hardware. Finally, the result analysis of the project is carried out and future scope of the project is discussed.
 VIRTUAL TOUCH SCREEN FOR ROBOTIC NAVIGATION
PROJECT GROUP: MR. PRANAM BORE, MR. KARTIK MOOLANI, MR. MAYANK GUPTA, MR. VIJAY WAGHMODE
Abstract: The main objective of the project is to develop object motion for wireless robot. It is a real time application. The project is based on the movement of hand or a finger on a touch screen & on this basis the robot also moves in the corresponding direction. The web-cam will capture the position of our hand or finger & with the help of image processing (using matlab) the distance covered will be calculated & viewed on the computer. This data will be transmitted by the RF transmitter via antenna & will be received by the receiver section. Here the information will be decoded by the decoder & as per the information the microcontroller will control the movement of robot by using the dc motor.
The block diagram consists of a transmitter and a receiver section. We are interfacing a web camera with the computer and using its vision area for detection Depending on the code generated by the program the receiver section responds.
The receiver consists of a modem and a decoder block that decodes the code and generates a corresponding signal code for the microcontroller. The microcontroller generates signal and drives the motor driver circuit. Depending on the direction of current the motor rotates in the corresponding direction.
 3D FACE RECOGNITION
PROJECT GROUP: ABHIJIT BHANDARI, PRAVIN AGRAWAL, ROHIT JADHAV, SUNIL GOLHAR
GUIDE: PROF.K.T. TALELE
Abstract: The main objective of the project is to develop a security system based on 3D human face recognition. The proposed system must recognize a 3D face model converted to a 2D image after its comparison with set of 2D images in the database.
We have used 3D Normalization and Template Matching for Recognition in our project, in which we compare the contour of the image (test image) with the set of images (training images), in the database.
The proposed system performs two important tasks i.e.”3D Normalization” and “Recognition”. 3D Normalization is rotation of a human face model along the three axis x-y-z in order to obtain frontal view of face model. This helps us to compare the input image with those in the database. This in turn increases the recognition rate. Recognition is performed to reveal the identity of the person whether he is present in the database or no. Such systems compare the portrait of a tested person with photos of people who have access permissions to object.
 HAND HELD FACE RECOGNITION DEVICE
PROJECT GROUP: DHAVAL KUNTE, VISHWESH SHANBHAG. SHAHU VIDHALE, PALAK KUBADIA
GUIDE: DR.S.T. GANDHE
Abstract: The aim of the project is to create a system which allows face recognition to be done by your very own mobile device instantly. In this world with the ever increasing security problems it is necessary to create a security system that is both portable and effective.
We use a 2D correlation technique for face recognition purpose. Image is clicked from a simple cell phone via a java applet and sent to the server. At the server end as soon as the image is received, automatically the face recognition program is executed and the desired result is obtained in the form of a text file saying whether a face has been recognised or not. This information is then relayed back to the mobile device via Bluetooth. The entire process occurs at its own without any human interference.
In this project we also learn about the different applications it can cater too and its scope in this present terror struck world.
 VIDEO STEGANOGRAPHY
PROJECT GROUP: MR. ADITYA KHUNE, MR. PANGERTOSHI IMSONG, MR. SUMEET PATIL, MR. HITESH CHAUDHARI
GUIDE: DR.S.T. GANDHE
Abstract: Steganography in the last few years has gained a wider audience due in part to the suspicion that the technology may have been used by terrorists to communicate plans for upcoming attacks. While those claims have never been formally substantiated, the technology has certainly been the topic of widespread discussion among the IT community and has provided the benefit of helping more people understand steganography and how it can be used today to conceal information.
Steganography is the practice of hiding private or sensitive information within something that appears to be nothing out of the usual. Think of steganography as meta-encryption: While encryption protects messages from being read by unauthorized parties, steganography lets the sender conceal the fact he has even sent a message. Steganography in Videos is similar to that of Steganography in Images, apart from information is hidden in each frame of video.
In this project, we propose a new video watermarking algorithm based on various Video Processing techniques. Different techniques are currently used to transfer messages secretly. In the past, messages could be identified very easily but in the present situation messages can be hidden in image, audio or in video files, which make the hidden text difficult to identify. This method is implemented in different fields such as army and Industries. Even personal data can be transmitted in secret by an individual. By using the lossless steganography technique messages can be send and receive safely.
 GPS BASED TRAIN TRACKING SYSTEM
PROJECT GROUP: PRIYA ANCHAN, ADITI GHAG, SHRUTI KHANDEKAR, TANVI RATNA PRASHANT
GUIDE: DR.Y.S. RAO
Abstract: Averting train accidents in India are a Herculean task as the Indian Railways runs a very large operation. Monitoring the performance of each train on a continuous basis is necessary to reduce risks. Thus we have decided to develop a train-tracking system that enables an effective way of monitoring the location and direction of movement of every train within a certain geographical boundary. The basic purpose of the project is to advance the train system technologically and make it safer. To achieve this we use three modules GPS, MICROCONTROLLER and GSM module.GPS (Global Positioning System) is used to determine location of the train. This information is transferred to station via GSM. GSM is used as channel for communication. In station the operators have privileges to see train location on a map with longitude and latitude.
 COMMUNICATION SYSTEM FOR ENERGY-METERS AND REMOTE SERVER
PROJECT GROUP: ASHWINKUMAR V. MADIVAL, KIRANKUMAR S. ITTAGONI, RAHUL P. BAVASKAR, YOGESH Y. JADHAV
GUIDE: Prof.N.A. BHAGAT
Co-GUIDE: Dr.Y.S. RAO
Abstract: The communication system for energy-meters and remote system i.e. Automatic Energy Meter System provides a cost effective, reliable & interference free data transfer between remote meter reading units & the convergence node. The meter reading & management processes are free from human involvement. Based on exiting telephone networks, it is very flexible for the utility companies to access, service and maintain this meter reading system. A user friendly and window based is designed which fully utilizes the personal computer‘s terminal and usart to achieve communication between the remote meter reading units and remote server.
Power utility companies have suffered revenue losses due to illegal electrical usage and uncollected bills for several years. Automatic remote meter reading system along with power line communication has been identified as a solution. Automatic meter reading data possess unique characteristic and the network needs innovative technology to record, monitor process the data. If this system via network communication is set in a power delivery system, a detection system for illegal usage may be easily added in the existing system. In the detection system, the second digital energy meter is used and the value of energy is stored. The recorded energy is compared with the value at the main kWh meter. In the case of difference between two recorded energy data, an error signal is generated and transmitter via network.
 RFID BASED AUTOMATION AND SECURITY SYSTEM
PROJECT GROUP: SHRENIK S. MHATRE, SWARAJ G. BANDGAR, SMIT K. DEDHIA
GUIDE: PROF.K.T. TALELE
Abstract: At the onset, the goal of our project was to design a security and automation system.We choose Radio Frequency Identification (RFID) ,an emerging technology applied for tracking and communication. RFID is an area of automatic identification that has quickly been gaining momentum in recent years and has now been seen as a radical means of enhancing the data handling process.RFID finds a wide range of applications than the conventional identification system like barcodes.
Radio-frequency identification (RFID) is an automatic identification method, relying on storing and remotely retrieving data using devices called RFID tags or transponders.Data within the tag can provide identification for an item in manufacture,goods in transit,location and identity of a vehicle or animal or individual.We can also store additional data to give some more specifications about the object. We chose the radio frequency identification technology as means of identifying individuals and this aspect can be used to implement a very secure and foolproof automation system.
RFID tags are available in two types, passive and active.We have used passive tags and reader to implement automation system.The information contained in the RFID card is read by the reader and it is passed on to the microprocessor Atmega 8535 for processing. Based on the information contained in the card the necessary applications are activated.
The other type of cards available are the active tags.These tags can be programmed as per user’s requirement.Accordingly we can design an encryption algorithm to read the data from these tags. This provides a higher level of security and can be implemented in transmission of secured data in military and other related application.
 MICRO-CONTROLLER BASED INVERTER USING SOLAR ENERGY
PROJECT GROUP: SHRINIVAS KALEWAD, AKASH JAUHARI
Abstract: Sun radiates enormous energy in form of heat and light. The light energy which is en-capsule in photons is a source of clean and cheap energy. Here is an attempt to utilize this energy for house hold applications.
Designing a Inverter to convert D.C power into A.C power. The Inverter consist of a switching device, circuit to commutate it and for its protection. Conventional inverters or analog inverters are inflexible i.e. their performance is fixed and cannot be altered. Our attempt is to design a more useful and efficient inverter. A digital inverter is one which allows us to control the output gain and power through use of a micro-controller. Thus to device a mechanism in which we control the switching action of MOSFET and its frequency of action.The complete system is formulated to charge through the day and then produces a power of 300 W. This will drive two tubes light and a fan. Attempt has to be made to achieve high conversion efficiency of converting first photo voltaic power into DC power, and secondly to convert DC -12 V into AC power of 230 V –rms.
 SAMPLING OF OBJECTS THROUGH IMAGE PROCESSING
PROJECT GROUP: MS.SONAL TELI, MS. PEMA CHOZOM THONGON, MR. KAUSHAL AJUGIA
GUIDE: PROF.R.G. SUTAR
Abstract: The aim of this project is to build a robotic arm that will be able to pick up the object of correct colour placed in the conveyor belt. Object will be placed on the conveyor belt, then the camera mounted exactly above the position of the object will take the image of the object, then the image will be processed in the computer using the program written in MATLAB. If the colour of the object matches the intensity of the required colour then the conveyor belt will move forward and drop the object at the end of the belt, but if the object does not match with the required colour then the conveyor belt will move forward and stop near the robotic arm then the robotic arm will pick it up and place it aside. We have chosen MATLAB since it has direct functions for image analysis. Also we have used ATMEL89C2051 along with all other hardware to control the movement of the robotic arm. The project involves simulation of the 89C2051 program in Kiel, interfacing of microcontroller with PC and operation of all the hardware components. The report details the results observed during the running the program and proper functioning of the hardware. Finally, the result analysis of the project is carried out and future scope of the project is discussed.
 TEXT EXTRACTION FROM IMAGE
PROJECT GROUP: MADHAVI SANE, ABHISHEK SHARMA, APURVA VANJARI
GUIDE: Dr.S.T. GANDHE
Abstract: The use of images to store and convey information is increasing daily. Text extraction from image, more commonly referred as Optical Character Recognition (OCR), involves recognition of printed text contained in an image. The concept of OCR is used in various applications like Automatic license plate detection, PDF to text convertors, Automatic cheque clearing and many more.
Various techniques are available for text extraction from images. The initial proposal of this project was to use only artificial neural networks for recognition of English alphabets of different font types and font sizes. However, we have implemented OCR using different algorithms, thus creating a comparative study of the techniques.
In this report, we will discuss some traditional methods like template matching and neural networks that are employed by the more advanced and artificially intelligent systems. Though neural networks provide better accuracy, it occurs at the cost of more processing time. Lately, non intelligent systems like Learning Vector Quantization and K-PCA have also shown to produce good results.
 SUSPICIOUS OBJECT DETECTION AND TRACKING
PROJECT GROUP: BHOOMIKA GALA, ABHISHEK MEHTA, PRATIK MEHTA, VINEET PARKAR
GUIDE: PROF.K.T. TALELE
Co-GUIDE: PROF.Y.S. RAO
Abstract: Given an image, object detection is to determine whether or not the specified object is present, and, if present, determine the locations and sizes of each object”.
The research for object detection and recognition is focusing on
- Representation: How to represent an object,
- Learning: Machine Learning algorithms to learn the common property of a class of objects
- Recognition: Identify the object in an image using models learned from 2).
This method is normally implemented by first picking out a part of the search image to use as a template: We will call the search image S(x, y), where (x, y) represent the coordinates of each pixel in the search image. We will call the template T(x t, y t), where (xt, yt) represent the coordinates of each pixel in the template. We then simply move the center (or the origin) of the template T(x t, y t) over each (x, y) point in the search image and calculate the sum of products between the coefficients in S(x, y) and T(xt, yt) over the whole area spanned by the template. As all possible positions of the template with respect to the search image are considered, the position with the highest score is the best position. This method is sometimes referred to as ‘Linear Spatial Filtering’ and the template is called a filter mask.
 POWER ANALYZER BASED ON LABVIEW
PROJECT GROUP: PRAJAKTA BHAGWATWAR, ANSHUL AGRAWAL, CHIRAG DARJI, AALAPEE PATEL
GUIDE: PROF.P.V. KASAMBE
Abstract: In this project, a PC based virtual instrument (VI) for power analyzer has been developed. The functions of power analyzer include the harmonic analysis and power measurements. The proposed VI is based on the LabVIEW (Laboratory Virtual Instrument Engineering Workbench) language under the workbench of Windows XP.
The real time inputs (voltage and current) are taken from the external circuit with the help of the analog channels on the NI ELVIS kit and the DAQ (Data Acquisition Card PCI-6251). These devices are used with LabVIEW to take input from the external circuit. This input can also be obtained from the serial port by using any microcontroller circuit.
The harmonic analyzer includes the output such as the harmonic spectrum, auto power spectrum, system frequency, harmonic components and the %THD (total harmonic distortion). The power measurements include instantaneous power, average power, apparent power, reactive power, power factor and power angle. The VI also logs the data into an excel sheet each time the program runs. This data is used to draw bar graphs to compare the above parameters at different instants of time at which the data is logged.
 FACE RECOGNITION IN VIDEOS
PROJECT GROUP: BHARGAV ASHTEKAR, REHAN SAYED, ANSH MAKHIJA, NEHA BORKAR
GUIDE: Prof.K.T. TALELE
Abstract: Given a video, face recognition is to determine whether or not the specified person is in the video.
The research for face detection and recognition is focusing on Video to Frame Conversion: To sample video at 15 frames per second. Face Detection: To detect the faces in the frames Face Recognition: To identify the person in the video. The video is taken in .avi or .mpg format thereby calculating the total no. of frames present in a video and then sampling it at a predefined rate. We develop a technique to identify an unknown person in an video image by using SMQT (Successive Mean Quantization Transform) and SNOW classifier and Template matching.
SMQT is used to extract features from the local area of an image. The SMQT uses an approach that performs an automatic structural breakdown of information. These properties will be employed on local areas in an image to extract illumination insensitive features.
This method is normally implemented by first picking out a part of the search image to use as a template: We will call the search image S(x, y), where (x, y) represent the coordinates of each pixel in the search image. We will call the template T(xt, yt), where (xt, yt) represent the coordinates of each pixel in the template. We then simply move the center (or the origin) of the template T(x t, y t) over each (x, y) point in the search image and calculate the sum of products between the coefficients in S(x, y) and T(xt, yt) over the whole area spanned by the template. As all possible positions of the template with respect to the search image are considered, the position with the highest score is the best position. This method is sometimes referred to as ‘Linear Spatial Filtering’ and the template is called a filter mask.
The system has 100% detection rate and 97-98% Recognition rate.
 SPEAKER RECOGNITION USING TMS320C6713DSK
PROJECT GROUP: SNEHA HEGDE, AMRUTA PENDHARKAR, PRATHAMESH PEWEKAR, ANIRUDDHA SATOSKAR
GUIDE: PROF.K.T. TALALE
Abstract: The technology of the automatic speech recognition is in full grow, a multitude of algorithms have been developed to improve the performance and robustness of ASR (Automatic Speech Recognition) systems. Automatic Speech recognition systems are increasingly widespread and used in very different acoustic conditions, and by very different speakers. The use of Mel frequency cepstral coefficients (MFCCs) for music information retrieval is one of the standard methods used in ASR systems. This paper describes a method to generate and process the Speech signal in digital domain using Texas Instruments’ TMS320C6713 DSK.
 AUTOMATED FOOD IRRADIATION AND PACKAGING
PROJECT GROUP: CLIFTON TUSCANO, KOMAL KAMPASI, PRIYANKA SUD, TUESAY SINGH
GUIDE: PROF.R.G. SUTAR
Abstract: The aim of the project is to develop a simulation for the Control System used in Food Irradiation Industries and to make a basic prototype of the whole process. The Programmable Logic Control is used as a controlling system because it is a very good benchmark for the automation of any electromechanical process. The project also involves the programming of the PLC used. The programs developed for machine control are stored in its non-volatile memory. The ROM memory contains the program information that allows the CPU to interpret and act on the Ladder Logic program stored in the RAM memory. The report details the results observed during the running of the controlling algorithm. The various parameters are controlled through the simulation to achieve a fully automatic, flawless and accident free control of the whole process. Finally the result analysis is carried out and the future scope of the project is discussed.
 DIGITAL AUDIO WATERMARKING SYSTEM
PROJECT GROUP: NARESH UNIYAL, SHREYA SHETE, NIKHIL DHOOT
GUIDE: PROF.N.A. BHAGAT
Abstract: Digital audio watermarking system is a technique for embedding additional data along with audio signal. Embedded data is used for copyright owner identification. A number of audio watermarking techniques are proposed. These techniques exploit different ways in order to embed a robust watermark and to maintain the original audio signal fidelity. This paper makes a tutorial in general digital watermarking principles and focus on describing digital audio watermarking techniques. These techniques are classified according to the domain where the watermark is embedded. A novel scheme that is able to merge digital watermarking and content authentication of digital audio is presented in this report. The embedding of additional data is performed in different signal domains. Watermark embedding is made by frequency hopping method in the Fourier domain, while the additional authentication data is hidden using the LSB modulation in the wavelet domain. The perceptual transparency is achieved using the frequency masking property of the HAS. The scheme obtains high robustness against standard watermark attacks and localizes accurately tampered parts of the audio clip.
 PC BASED MULTI-TOUCH INTERFACE SYSTEM
PROJECT GROUP: PRERAK H. DESAI, LABHESH HASE, PALLAV EDWANKAR, MIHIR JOSHI
GUIDE: PROF. K.T. TALALE
Abstract: Technology has evolved to a great extent and it promises to evolve further in the years to come. For instance, we have faster processors that can execute thousands of instruction within a second. These developments haven’t stopped and scientist, developers and companies around the globe are continuously working on ways to improve the technology in every possible way. Apart from speed, throughput and various other aspect of a computer there is one other important factor that is to be taken into account when designing a system. This aspect is the quality of interaction that takes place between user and computer. There are various ways in which quality of interaction can be improved.
The project aims at using one such low cost interaction method, a multi touch system, to show how the interaction between human and machine can be improved. This improved interaction is shown by controlling a microcontroller based robot according to the touch input given by the user.
The touch screen is not capacitive based screens that are used in the mobile phone these days. But the touch screen is implemented using Infra red LED and an Infra red camera.
The robot is made using DC motors and a microcontroller. The feedback includes images captured by the over head camera. The communication is done using wireless module that implements IEEE 802.15.4 Zigbee protocol. The programming is done on a Java based platform and C language.
This project can be used in scenarios wherein the operator has to control multiple objects simultaneously and the environment cannot be viewed directly by the operator.