Electrical and Electronics Commons^™

Two-Dimensional Quantum Material Identification Via Self-Attention And Soft-Labeling In Deep Learning, Xuan Bac Nguyen, Apoorva Bisht, Benjamin Thompson, Hugh O.H. Churchill, Khoa Luu, Samee U. Khan

Electrical Engineering and Computer Science Faculty Publications and Presentations

Detecting two-dimensional (2D) materials in silicon chips presents a significant challenge in the field of quantum machines due to the difficulty of data collection. Specifically, among thousands of flakes, not all flakes are useful or well-annotated, resulting in noisy and hard samples within the dataset, which challenges the deep neural network (DNN) to learn. To address this problem, we propose a novel method for identifying quantum 2D flakes even when there is a high rate of missing annotations in the input images. In particular, we first propose a new mechanism for automatically detecting false negative flakes that are missing annotations. …

Automatic Testing Of Organic Strain Gauge Tactile Sensors., Brian P. Goulet

Electronic Theses and Dissertations

Human-Robot Interaction is a developing field of science, that is posed to augment everything we do in life. Skin sensors that can detect touch, temperature, distance, and other physical interaction parameters at the human-robot interface are very important to enhancing the collaboration between humans and machines. As such, these sensors must be efficiently tested and characterized to give accurate feedback from the sensor to the robot. The objective of this work is to create a diversified software testing suite that removes as much human intervention as possible. The tests and methodology discussed here provide multiple realistic scenarios that the sensors …

Magic: The Gathering Card Virtualizer, Vincent Garbonick, Jacen C. Conlan, Jaret A. Varn

Williams Honors College, Honors Research Projects

Any well-versed Magic: The Gathering (MTG) player or collector knows how difficult it can be to keep track of all cards in their collection. Some spend hours searching for that one specific card, and others are constantly scouring the internet for how much their collection costs. However, this issue does not only affect casual fans. Resale companies spend hours a day determining the costs of cards, and tournament judges painstakingly check players’ decks to ensure they are not cheating. To assist with these struggles, the design team proposed to create the MTG Card Virtualizer. This device scans MTG playing cards …

A Brief Bibliometric Survey On Night Vision Bot Using Dynamic Ir And Object Detection, Devesh Abhyankar Mr., Gurumoorty Suresh Mr., Hrithik Sambhaji Karjule Mr., Parth Bhardwaj Mr., Harish Muleva Mr., Anurag Mahajan Dr.

Library Philosophy and Practice (e-journal)

This study aims to analyse the work done in the field of Night Vision Robots using IR and Object Detection from 2011 to 2021, using the bibliometric methods. This paper presents a Scopus database review on "Night Vision Bot using Dynamic IR and Object Detection". The necessity for doing this bibliometric survey is that to know how the technology in the field of mobile robotics and night vision, as well as to object detection, has evolved over the years. This paper shows the importance of Night Vision Robot from the year 2011 and continued up to 2021 April. The database …

Collision Avoidance Smartphone, Matt Columbres, Aaron Parisi, Joey Schnecker, Luis Wong

Electrical Engineering

There are many instances in day-to-day life where people cannot or would rather not pay full attention to their surroundings. Walking while preoccupied with a smartphone or walking while blind are excellent examples where technology could be used to make the task of avoiding 2collisions reactive, instead of proactive. A device which monitors a user’s surroundings and notifies the user when a potential collision is detected (and, additionally, notifying them as to where the obstacle is with respect to them) could be used to make walking distracted less of a hazard for the user and those around the user and …

From A Locally Competitive Algorithm To Sensory Relevance Models, Walter Woods

Electrical and Computer Engineering PhD Day

This poster addresses the development of a new Machine Learning (ML) mechanism, the Sensory Relevance Model (SRM), as a means of splitting information processing tasks into two sub-tasks with more intuitive properties. Specifically, SRMs are a front-end for other ML techniques, re-mapping the input data to a similar space with significantly more sparsity, achieved through the transformation and suppression of inputs irrelevant to the task. Prior work has attempted to reveal this information for Neural Networks (NNs) either as a post-processing step via saliency maps or through a simple masking of the input achieved with a dot product (so-called ``attention'' …

Team Omnimouse, Derek J. Halman, Josh B. Porter, Steven A. Silver, Ian S. Stemper

Computer Engineering

INFORMATION, DATA, FIGURES AND DRAWINGS EMBODIED IN THIS DOCUMENT ARE STRICTLY CONFIDENTIAL AND ARE SUPPLIED ON THE UNDERSTANDING THAT THEY WILL NOT BE DISCLOSED TO THIRD PARTIES WITHOUT THE PRIOR WRITTEN CONSENT OF QUALITY OF LIFE PLUS.

Visual Geo-Localization And Location-Aware Image Understanding, Amir Roshan Zamir

Electronic Theses and Dissertations

Geo-localization is the problem of discovering the location where an image or video was captured. Recently, large scale geo-localization methods which are devised for ground-level imagery and employ techniques similar to image matching have attracted much interest. In these methods, given a reference dataset composed of geo-tagged images, the problem is to estimate the geo-location of a query by finding its matching reference images. In this dissertation, we address three questions central to geo-spatial analysis of ground-level imagery: 1) How to geo-localize images and videos captured at unknown locations? 2) How to refine the geo-location of already geo-tagged data? 3) …

Human Action Localization And Recognition In Unconstrained Videos, Hakan Boyraz

Electronic Theses and Dissertations

As imaging systems become ubiquitous, the ability to recognize human actions is becoming increasingly important. Just as in the object detection and recognition literature, action recognition can be roughly divided into classification tasks, where the goal is to classify a video according to the action depicted in the video, and detection tasks, where the goal is to detect and localize a human performing a particular action. A growing literature is demonstrating the benefits of localizing discriminative sub-regions of images and videos when performing recognition tasks. In this thesis, we address the action detection and recognition problems. Action detection in video …

Holistic Representations For Activities And Crowd Behaviors, Berkan Solmaz

Electronic Theses and Dissertations

In this dissertation, we address the problem of analyzing the activities of people in a variety of scenarios, this is commonly encountered in vision applications. The overarching goal is to devise new representations for the activities, in settings where individuals or a number of people may take a part in specific activities. Different types of activities can be performed by either an individual at the fine level or by several people constituting a crowd at the coarse level. We take into account the domain specific information for modeling these activities. The summary of the proposed solutions is presented in the …

Feature Pruning For Action Recognition In Complex Environment, Adarsh Nagaraja

Electronic Theses and Dissertations

A significant number of action recognition research efforts use spatio-temporal interest point detectors for feature extraction. Although the extracted features provide useful information for recognizing actions, a significant number of them contain irrelevant motion and background clutter. In many cases, the extracted features are included as is in the classification pipeline, and sophisticated noise removal techniques are subsequently used to alleviate their effect on classification. We introduce a new action database, created from the Weizmann database, that reveals a significant weakness in systems based on popular cuboid descriptors. Experiments show that introducing complex backgrounds, stationary or dynamic, into the video …

Electronic Image Stabilization For Mobile Robotic Vision Systems, Michael John Smith

Theses and Dissertations

When a camera is affixed on a dynamic mobile robot, image stabilization is the first step towards more complex analysis on the video feed. This thesis presents a novel electronic image stabilization (EIS) algorithm for small inexpensive highly dynamic mobile robotic platforms with onboard camera systems. The algorithm combines optical flow motion parameter estimation with angular rate data provided by a strapdown inertial measurement unit (IMU). A discrete Kalman filter in feedforward configuration is used for optimal fusion of the two data sources. Performance evaluations are conducted by a simulated video truth model (capturing the effects of image translation, rotation, …

Discovering The Merit Of The Wavelet Transform For Object Classification, Matthew D. Eyster

Theses and Dissertations

Vision is the primary sense by which most biological systems collect information about their environment. Computer vision is a branch of artificial intelligence concerned with endowing machines with the ability to understand images. Object recognition is a key part of machine vision with far reaching benefits ranging from target recognition, surveillance systems, to automation systems. Extraction of salient features from an image is one of the key steps in object recognition. Typically, geometric primitives are extracted from an image using local analysis. However, the wavelet transform provides a global approach with good locality. Additionally, the directional and multiresolution properties may …

Object Classfification In Computer Vision With Discriminant Analysis, Amir Hamzahan

Makara Journal of Technology

A robotic sensor system is always supported by a computer system called ‘computer vision’. The important concept of computer vision is object classfifi cation. In this study two algorithms for object classifi cation in this system will be compared. Firstly, A simple method that do not need complex computation and that considered as an informal method is called binary tree decision structure. This method is based on modest caracteristic decriptors of an object such as vertical line, horizontal line or ellipse line. Unfortunately this method has weakness in recognize an image that contaminated by a noise. Secondly, a more formal …

Adaptive Color Correlation Of Knots In Wood Images And Weighted-Value Product Selection Methods In A Machine Vision System, John Robert Goulding

Dissertations and Theses

The biggest obstacle to robust color image processing of wood is in developing a color model that represents all possible defect colors. When the color model is too general or too specific, defect recognition fails because too many or too few non-defect pixels match the model, respectively. Because a color image of wood contains far more clear and clear-grain colored pixels than grain-knot and knot colored pixels, it is beneficial to first statistically identify and remove the clear and clear-grain colors and to use the accumulated data to simultaneously enhance and normalize the remaining grainknot and knot colored pixels. This …

An Implementation Of A Versatile Camera Calibration Technique For High-Accuracy 3d Machine Vision Metrology Using Off-The-Shelf Tv Camera And Lenses, Bolang Li

Theses

This thesis studies and implements a new versatile camera calibration technique for high-accuracy 3D machine vision metrology using off-the-shelf TV camera and lenses developed by Roger Tsai [1]. This technique builds up a unique relationship from the world coordinate system to the computer image coordinate system of calibration points by using a radial alignment constraint. The technique has advantage in terms of accuracy, speed, and versatility over existing techniques.

The fundamental knowledge for using this technique is presented in this thesis first, followed by an overview of the existing calibration techniques, and a detailed description of the new technique. The …

Position Recovery And Motion Analysis From Stereo Image Sequences, Yuxin Xie

Theses

Based on four stereo images, a new concept of unified optical flow field has been developed recently [3]. It is a generalization of the optical flow in stereo imagery. A set of fundamental equations is established to characterize the unified optical flow field. Another set of equations is then derived from which 3-D structure can be reconstructed and its motion can be analyzed by the use of the unified optical flow field quantities. In this thesis, we use some synthetic images of a rotated conical structure to study and test the theoretical results. The simulation and results are presented, thus …

Studying The Mechanical Response Of Outer Hair Cells To Electric Pulses Using Image Sensors, Wayel F. Mutlag

Theses

The purpose of this research was to develop a way to improve the method used by Dr. Santos-Sacchi at the University of Medicine and Dentistry of New Jersey to detect the motion of the outer hair cell (OHC). As part of the work, the construction and functioning of the OHC are presented and two designs to detect the motion of the boundary of the OHC were investigated.

The first design utilizes the position sensitive detector (PSD) S3932 made by Hamamatsu Inc. This detector, however, required excessive intensity of the optical signal that was not compatible with the operation of the …

Image Analysis Using Gabor Transforms: Correlated With Human Saccadic Movement Data To Identify The Human Visual Search Strategy, Ralph J. St John

Theses and Dissertations

This thesis investigated whether a relationship exists between the Gabor Correlation Coefficient (GCC) magnitudes of fixation points for a set of images viewed by six subjects and the human visual search strategy. A couple of different relationships were investigated. First, the data were analyzed to determine if the GCC magnitudes of human fixation points predicted the time ordered sequence of human fixation points during the performance of a visual search. Second, the data were analyzed to determine if there was a significant difference between the GCC magnitudes of the human fixation points and the GCC magnitudes of a random set …

3-D Structure Recovery Using Unified Optical Flow Field Approach, Anees Ahmad

Theses

Recovery of the 3-D structure that is characterized by an N^th order factorable and/or non-factorable polynomial equation is studied in this thesis. Analytical tools have been employed to analyze and solve the problem. The solution is based on the direct method, which is derived from the unified optical flow field, which, in turn, is an extension of optical flow to spatial image sequences. Least squares formulation has been employed for optimum estimation. The method used does not require the establishment of point to point correspondence, nor the estimation of the optical flow, as an intermediate step.

The theoretical results …

Automatic Infrared And Visual Inspection Of Solder Joints On Pcbs, Jiang Huang

Theses

This thesis reports the research results of automatic inspection of solder joints on printed circuit boards (PCBs). The previous work on this regard has been advanced significantly in the following three aspects. With the application of some microlens and the principle of geometric optics solder joints of real size are inspected in this work instead of larger simulation solder joints in the previous work. A new set of features is formulated and used for infrared inspection. Consequently, better results have been achieved in this work.

Reasonably good results demonstrate the capability of infrared inspection of solder joints on PCBs. Once …

Direct Recovering Of Composite Surface Using Uoff, Zhesheng Huang

Theses

The research work reported in the thesis is motivated by the problem raised in the computer vision area. It utilizes direct method to recover surface structure of objects in 3-D space from a pair of stereo images. The direct method, compared with optical flow-based approach and feature-based approach, is robust and computationally efficient. However it can only recover planar surface structure so far. A new approach to recover structure from a pair of stereo images based on unified optical flow field (UOFF) is developed recently. It can recover curved surface structure.

In this thesis, the problem of recovering surface structure …

Analysis And Implementation Of A Navigation System Using Vanishing Points In A Generalized Environment, Rolf Schuster

Theses

The development of accurate sensors is of crucial importance in navigation of mobile autonomous robots. The following master's thesis analyzes the use of vanishing points for robot navigation. Parallel lines in the environment of the robot are used to compute vanishing points which serve as a reference for guiding a robot. To accomplish the navigation tasks, three subtasks are to be performed: detection of straight lines, computation of vanishing points, and robot navigation with vanishing points.

An edge detection algorithm is presented that combines Sobel and Laplacian of Gaussian operators. The algorithm preserves the precision of the Laplacian of Gaussian …

A Machine Vision System With Remote Host, Chieyeon Jason Chen

Theses

The modem plays an important role in data communication using the Public Swiched Telephone Network. In this thesis, the modem function is merged into a low cost machine vision system based on the Motorola 68HC11. Evalation Board The system comprises the intelligent machine vision con- troller and a remotely located host computer, such as an IBM-PC. It, supports extra long distance communication capability so that the controller and the host can be anywhere in the world where then is a telephone line. Because of the limit of a 2100 bps modem, soft ware data compression is implemented here to save …

Estimating 3-D Motion And Structure Parameters Of A Rigid Planar Object By Establishing Landmark Correspondences, Herman Agustiawan

Theses

The application of a landmark based approach in recognizing 3-D rigid planar objects[3] along with two different motion and structure estimation algorithm [21][23] have been studied to perform motion and structure estimation tasks of moving objects. The recognition algorithm is based on a sphericity value derived from affine transformation that maps feature points in the object under observation from first to second view in the 2-D image space. The motion and structure estimation algorithm is based on a unique mapping given four or more feature point correspondences.

More than ten experiments have been done including small and large rigid motions. …

Recovering Surface Structure Using Direct Method, Zhu Ying

Theses

A new approach to recovering surface structure from a pair of stereo images is proposed in this thesis. The approach is based on an extension of the concept of optical flow to a spatial sequence of images. z.e., the unified optical field (DOFF). A direct method in spatial domain is developed which is able to recover surface structure that can be characterized by an Nth order polynomial equation. The order N of the surface can be one. a planar surface. or two, a cylindrical or spherical surface. for example. or any number higher than two. This is a big improvement …

A 512 X 512 Random Addressable Variable Resolution Image Sensor, Chakravarthy Yarlagadda

Theses

This thesis presents the design and layout of a 512 x 512-element with 20μm x 20μm pixels image sensor that emulates the properties of the human eye and is useful in high speed processing of visual data for machine vision applications. The imager will be able to operate with random-address selection in variable resolution or with multiple regions of interest to reduce the number of pixels readout. The selection of the pixels is done by x-y addressing the MOS switches which connect the photodiode pixel to the output through the sense lines. The design of column decoder and row decoder …

Landmark-Based Partial Shape Recognition By A Two-Stage Bam Neural Network, Xianjun Liu

Theses

In this thesis, we develop a Bitfirectional Associative Memory (BAM) based neural network to achieve partial shape recognition. To recognize objects which are partially occluded, we represent each object by a set of landmarks. The landmarks of an object are points of interest relative to the object that have important shape attributes. To achieve recognition, feature values (landmark values) of each model object are trained and stored in the network. Each memory cell is trained to store landmark values of a model object for all possible positions. Given a scene which may consist of several objects, landmarks in the scene …

Microcomputer-Based Visual Controller And Data Logger System, Xiao-Fang Qi

Theses

This thesis paper describes the firmware design, instead of the hardware logic design, makes the machine vision hardware system economical, intelligent and expandable. It shows the construction and testing of a novel remote low speed machine vision control system, which also has functions as data logger and is capable of auto control. The system has 9 analog input channels available, and more. It also has 4 digital output channels for controlling external devices. This work also implemented a way to transfer the grabbed image to a PC/XT. In the 256 graphics mode of the VGA, 64 B&W gray levels can …

Landmark-Based Shape Recognition By A Modified Hopfield Neural Network, Kuowei Li

Theses

In this thesis, we introduce a new thethod to achieve partial shape recognition by means of a modified Hopfield neural network. To recognize partially visible object, we represent each object by a set of "landmarks." The landmarks of an object are points of interest relative to the object that have important shape attributes. Given a scene consisting of partially occluded objects, a model object in the scene is hypothesized by matching the landmarks of the model with those in the scene. A measure of similarity between two landmarks, one from a model and the other from the scene, is needed …