DOWNLOAD FREE PROJECTS | ENGINEERING NOTES | PLACEMENT NOTES | MATLAB AND ANDROID PROJECTS

Robust Phase Retrieval with Prior  Knowledge Output  1 Introduction In many imaging applications, one can only measure the power spectral den-sity (the magnitude of the Fourier transform) and not the phase of the signalof interest. For example, in optical settings, devices like CCD cameras only  capture the photon flux, so important information stored in the phase is lost.The challenge is to reconstruct the original signal from measurements of its Fourier magnitude. This problem, known as phase retrieval, arises in many scientific and engineering applications, including optics, crystallography, and astronomy. 2 Problem Description Suppose x = (x1, . . . , xN ) is a non-negative signal, and denote by F : x → F the operator that returns its N-point discrete Fourier transform (DFT). We observe b = (b1, . . . , bN ) ∈ R N+ , the magnitude of the Fourier transform of the signal, and wish to find x that satisfies |Fx| = b. Phase retrieval can be posed as the optimiz

class book cover recognition OUTPUT 1. Introduction: The goal of our project is to detect weapons’ positions that concealed underneath peo-ple’s clothes and at the same time to have better visual performance on the infused colorimages. 2. Project description According to statistics, 13,286 people were killed and 26,819 people were injured in the US in 2015. In order to effectively prevent the occurrence of firearms incidents, hid- den weapon detection technology (Concealed Weapon Detection, CWD) has become an important research topic. Researchers have been using image fusion technique to combine the weapon information(from infrared imaging) and scene information(from visible light imaging). Here we aim at improving the fusion of IR image and RGB image in order to pinpoint the position of the weapons. We also aim at firearm shape recognition in order to have better knowledge of the concealed weapon. 3. Implements This project will use Matlab and may use Caffe framewor

Vehicle Driver Assistant System OUTPUT 1. Introduction The driver assistant system has been implemented in many vehicles in the market to help   protect both the safety of drivers and pedestrian. The related technologies are also used in   developing the autonomous vehicles. Using the image processing techniques, we plan to   build a vehicle driver assistant system based on MATLAB to provide drivers reliable  analysis of the traffic information and assist the driver to identify the car information  ahead including the vehicle manufacturer and the name of model. 2. Goal Our goal is to use the video image processing techniques to build a robust system to   achieve several functions below. - Determine if the car is driving stable on the lane. If the car moves toward another lane  without turning on the sign signal, there will be a warning to the driver as a reminder. - Use the video image of the back of front car to determine car manufacturer and model. - Traffic

Anomaly detection in optical floating zone single crystal growth Output  I. BACKGROUND AND MOTIVATION A vital component to the rise of the semiconductor in-  dustry is the viability of sizable high quality single crystal silicon, which is predominantly obtained via Czochralski  or the floating zone method 1 . The latter is also widely  employed in many cutting-edge materials research includ-  in novel superconductors and artificial ruby/sapphire  synthesis. In this method, the precursor mixture is  pressed into a rod, melted at one end (usually above 1000  degree Celsius), and the heating source sweeps the molten  zone slowly to the other end to enable high quality crystal  formation in the re-crystalized region.  The the stability of the molten zone is thus of paramount significance in the entire process, whose main-  tenance is often empirical driven and demands 24/7 at- tendance of manual intervention, especially in the early  stage of new materials development. Recen

Recognition and Translation of Thai Characters and Words from Text OUTPUT Goal/Objective: ● Detect, segment, and translate Thai characters from images of printed text. (Character classification) ○ This includes characters that have been rotated, slanted, within noisy images, and in uneven lighting conditions. ○ Initially, we will test and work on detecting a single font of Thai, then add different fonts and sizes to improve algorithm robustness. ○ We will also just start with the 44 consonants and if we are able to characterize these, we will move on to the different vowels that exists on top, underneath, and on the side of consonants. ● We will collect data​ from Thai newspaper websites. ○ First our training data on one font would be done by copying text into google docs and printing them out as a jpeg images. ○ We will physically print them out and take photos of these document in different lightings and orientations. Use these as both training and test data. ○

Optical Verification of Mouse Event Accuracy OUTPUT Motivation: The project is designed to address the issue of fair gameplay in professional  competitive gaming events comparing point tracking techniques, assessing optical flow,  and multi-object tracking. In doing so, we aim to analyze and determine the most robust  system for accurately correlating mouse movements against their corresponding motion  on the player’s screen. By cross referencing these data points, we will be able to confirm the validity of a player’s movement during a gaming event.  Proposal  Optical verification of mouse events will track mouse accuracy by analyzing a marker on  the mouse and comparing, in real time, mouse movement to the resulting movement on  screen. We will add multiple visual markers on the mouse and then perform feature extraction techniques to find their position and calculate the displacement vector  between frames. We will compare multiple algorithms to ensure consistency in the

App for Evaluating Handwritten Mathematical Expressions OUTPUT OVERVIEW We propose to develop an Android app to recognize and evaluate handwritten mathematical expressions and equations. For our initial scope, we plan to focus on recognizing digits (0-9) and basic mathematical operators (+, -, x, and /), and displaying the computed result on the device’s viewfinder. This will be our primary goal. To accomplish this goal, we plan to explore different optical character recognition (OCR) algorithms, including the template matching algorithms we’ve seen in class. Since the app will run on a mobile device, our algorithm will need to tradeoff between performance and computational complexity. If time permits, we would have several potential stretch goals to choose from. For example, we may add augmented reality features to the app. It would be interesting to overlay answers, error corrections, or incremental calculations onto the device’s viewfinder. We may also be able to identif

Instant Camera Translation and Voicing Output Goals: Develop an Android app for signs translation from English to Chinese or other  foreign languages with auto voicing. Useful for sign recognition at airport or on  road, menu translation while dining at a restaurant, guide translation at tourist  attractions, text translation while reading a book. Help foreigners to guide and  enjoy themselves when travelling in the states without the hassle of opening a  digital dictionary and typing in the text. Sometimes, when typing we could easily  mistake one letter or another and we want to avoid any mistake by recognizing the  words directly. Just focus your camera and get the translation popping out  immediately! In addition, we will also test the performance of our app on the  database of signs. We’ll also conduct field test of the app at airports and inside  buildings. Plans :  Clean the image and segment the text part of signs. We’ll mainly focus on  sign texts which are s

 Sign Language Recognition on Android OUTPUT Project Idea and Goals For our project, we would like to develop an algorithm for human hand detection and in-terpretation. This means that we not only want to identify human hands in images, but also correctly model their pose and classify gestures. Although this algorithm has many varied applications such as human-machine interface commands or machine understanding of human behavior, we will use this algorithm specifically to detect and interpret American Sign Language gestures. Specific goals for the project include (1) live video detection and conversion of sign language to text in real time and (2) mobile implementation on an Android device. Proposed Methodology In order to do this idea justice, we plan to organize the project into an image processing component for hand segmentation and modeling of the gesture (for EE 368), and a machine learning component for hand pose interpretation (for CS 229). We would like the imag

Traffic Sign Recognition for Vision-based Driver Assistance in Land-based Vehicles OUTPUT: Motivations Autonomous vehicle has been an active area of research for a few decades. And  intensive research has been done on using a front viewing camera for vehicle  localization and navigation, environment mapping, and obstacle avoidance. In this  project, I want to look into various digital-image-processing algorithms for  detection and classification of traffic signs, and compare the performance under  different image qualities.  Objectives The main objective of this project is to successfully detect traffic signs from a  stationary image of a normal street scene. The algorithm should also be able to  extract features from the traffic signs and find the best match from predefined  templates. The algorithm should also be robust towards variation of the quality of  the images. Potential Challenges Depending on the quality of the images, several challenges can be a

Video creation from Random Video Frames output Goals - This project is motivated by the talk given by TJ Lane and the need to reconstruct randomly ordered frames into a coherent video. This algorithm would allow scientist to observe dynamic information of samples from many repeated measurements of the sample at various states. The aim is to reconstruct a movie whose frames have been randomly reordered/repeated, much like what would be observed in an experiment at SLAC. This technique has the potential of opening a new set of experiments that can be carried out, allowing researchers to capture how proteinschange structure even though the sample is destroyed during imaging. To approach this problem, I plan on first designing the algorithm using various videos and computer simulations of protein dynamics possibly obtained from TJ Lane. The core of the algorithm will focus on computing metrics describing the distance between frames to find what the most likely ordering is. Possib

Mobile Phone Book Finder Motivation :  When people are looking for a book in the library, the call number provided by the library catalog system can easily lead them to the correct bookshelf. However, people may have to spend a very long time to find the exact position of the book on that shelf. Not only do patrons not know what the binding looks like, but books are often reshelved incorrectly and make finding the book a hassle. Our group is proposing to create a mobile phone book finder which can find the book’s position by taking a picture of the shelf and knowing the image of the book spine from a database, point out on the phone where the desired book is located on the shelf. Implementation :  Determining the correct position of the desired book from the shelf picture would require accurate segmentation of the books, as well as a matching algorithm of the book in order to output the most likely position of the book. The photo of bookshelf would be subject to several

Automatic Cell Detection of Liver Tissue Section Image OUTPUT 1 Introduction The Nusse Lab of the Stanford Institute of Stem Cell Biology & Regenerative Medicine studies  the regenerative properties of the liver. The goal of this project is to help graduate students in the  Nusse Lab automate the tasks of cell counting and characterization of liver tissue section images,  leveraging image processing and machine learning techniques.  Currently, the cell counting tasks of tissue section images are done by hand, in a manual and  laborious manner, because general purpose image processing software such as Image J does not  adequately address the specific need for these types of images and there are no commercially available products solving this problem [Gri15]. While previous projects have dealt with cell  counting or characterization of cell culture images, this project tackles the more difficult problems  presented by tissue section images due to their non-homogeneous natu

 Facial Feature Detection and Changing on Android OUTPUT: Introduction and Background  One’s face directly expresses his/her emotions, but sometimes it would be interesting to add a little bit more comic element, such as the popular emoticons. I plan to develop an Android app that can detect facial features, and morph the facial features to match a certain emoticon. Such a feature could be potentially fun and useful for social networking apps, photo editing, gaming, etc. Using face and facial feature recognition in emotion extraction have been studied by many research papers [1], and appears interesting to me as well. However, I will focus on altering facial feature to simulate emotion rather than extracting emotion, because the latter would require prestoring and extracting features for each emotion, and is thus more complicated. Given the time scale of the project and computational ability of the mobile, I think the current goal is more realistic. Proposed Work and Timel

Mobile Address Tagging for OpenStreetMap OUTPUT: DESCRIPTION: The objective of this project is to create an Android smartphone application which can tag features in OpenStreetMap (OSM) using the phone’s camera and GPS. OpenStreetMap is a crowd-sourced mapping project, analogous to Wikipedia for cartography. While many features such as roads and buildings can be readily drawn from aerial and satellite imagery, a complete map requires that these objects be tagged with their names, addresses, and other data. This has traditionally been a tedious task requiring field notes and hours entering tags on a computer, but mobile devices have the potential to radically accelerate this process. Our application will allow the user to snap a photograph of a building’s address or a business’ sign, and text from the image will be detected, parsed and applied as tags to the appropriate object in OpenStreetMap. At a minimum, our application will correctly identify the address of the building i

Modifying Images for Color Blind Viewers OUTPUT: PROJECT DESCRIPTION: Color blindness affects roughly ten percent of human males. It is possible for a woman to be color blind, however, because the main form of color blindness manifests as a defect in the X chromosome, most color blind individuals are men. Of those diagnosed with color blindness, over ninety-nine percent of them suffer from some sort of red-green deficiency, where they are unable to distinguish well between red and green. The goal of this project is to design and implement an image processing algorithm to help red-green color blind individuals perceive color diversity in digital images. Because color blindness almost always refers to a deficiency of color perception rather than an absence of it, one method of correcting images for the red-green color blind is to enhance the contrast between red and green pixels. Another approach is to map both red and green to different colors easily distinguished by a

Detection of flood-survivors through IR imagery on an autonomous drone platform OUTPUT PROJECT PROPOSAL Objective The goal is to detect human survivors in infra-red and visual imagery to enable autonomous search and rescueoperations through drones. Goals 1. Demonstrate a proof-of-concept of the algorithms in MATLAB through recorded video sequences 2. Hardware integration and validation of the algorithm (written in C++) running real-time on the NVIDIA Jetson TK1 processor on a DJI M100 quadcopter. Solution The proposed solution will emulate the two-stage approach described by Teutsch et. al. [1]: (1) the application of Maximally Stable Extremal Regions (MSER) to detect hot spots and (2) the verification of the detected hot sportsusing a Discrete Cosine Transform (DCT) based descriptor and a modified Random Naive Bayes classifier. Project Outline Immediately after disasters such as floods in the watersheds of major rivers and affected areas of major hurricanes, th

Light field imaging to enable band OUTPUT OVERVIEW Our project aims to obscure and reveal hidden scenes and objects in a picture, based upon the principles of Moiré level lines and aliasing. Chossen & Hersch [1] propose the mathematics and examples of animated, beating shapes hidden in Moiré level lines based upon a simulated elevation profile of the object. Rather than virtually generate the scene to encode, we propose the use of a Kinect time-of-flight camera to gather the depth map of a scene in real time. Additionally, we’d like to explore the use of two LCD panels for the banding images. One would have the backlight removed and just be a transparency mask. At the demo & poster session, we would allow users to capture a scene with the camera and move the mask over the display to reveal the hidden scene. Here’s an example of the expected result: Here’s our proposed methodology: 1. Use a kinect to capture the depth map of an object we wish to obscure w

Android-Based Digital Image Steganography and Steganalysis OUTPUT OVERVIEW Steganography is the discipline concerned with achieving confidential communication by hiding information in plain sight. Media for hiding this information include images, video, audio and markup languages (Papapanagiotou et al. 2005). Steganographic schemes typically exploit information redundancies which are not easily perceptible. Digital images tend to exhibit such redundancy, and thus are a popular medium for steganography. Throughout the years, many different methodologies have been proposed in the literature, of which steganographic schemes like F5, OutGuess or Yet Another Steganographic Scheme (YASS) are a sample. While some steganographic techniques operate in the primal domain, the majority of newer techniques utilize the frequency domain to conceal information. F5,OutGuess and YASS occupy this group, embedding the information to hide in the image’s discrete cosine transform (DCT) coeffic

Automated Restyling of Human Portrait Based on Facial Expression Recognition and 3D Reconstruction OUTPUT 1.1 Depth Data Acquisition We plan to use Microsoft Kinect to capture half body portraits of the user. We use not only color but depth information for restyling purposes. 1.2 Facial Feature Extraction With the photos of facial expression collected as our dataset, we then extract the features from them for later model training. This process can be done in several ways, one involves filtering certain colors of the pictures to obtain the shape of facial features (such as mouth, eyes, and eyebrows), then calculating their angle, momentum, or other characteristics. The results can be used as our training features. Using Scale-invariant feature transform (SIFT) algorithm is another way to collect the feature descriptions.In this project, We plan to integrate existing approches and develope our own feature collecting stategy. The relevant research papers are listed b