Engineering Commons^™

Uncovering And Mitigating Spurious Features In Domain Generalization, Saeed Karimi, Hamdi̇ Di̇bekli̇oğlu

Turkish Journal of Electrical Engineering and Computer Sciences

Domain generalization (DG) techniques strive to attain the ability to generalize to an unfamiliar target domain solely based on training data originating from the source domains. Despite the increasing attention given to learning from multiple training domains through the application of various forms of invariance across those domains, the enhancements observed in comparison to ERM are nearly insignificant under specified evaluation rules. In this paper, we demonstrate that the disentanglement of spurious and invariant features is a challenging task in conventional training since ERM simply minimizes the loss and does not exploit invariance among domains. To address this issue, we …

Enhancing Water Safety: Exploring Recent Technological Approaches For Drowning Detection, Salman Jalalifar, Andrew Belford, Eila Erfani, Amir Razmjou, Rouzbeh Abbassi, Masoud Mohseni-Dargah, Mohsen Asadnia

Research outputs 2022 to 2026

Drowning poses a significant threat, resulting in unexpected injuries and fatalities. To promote water sports activities, it is crucial to develop surveillance systems that enhance safety around pools and waterways. This paper presents an overview of recent advancements in drowning detection, with a specific focus on image processing and sensor-based methods. Furthermore, the potential of artificial intelligence (AI), machine learning algorithms (MLAs), and robotics technology in this field is explored. The review examines the technological challenges, benefits, and drawbacks associated with these approaches. The findings reveal that image processing and sensor-based technologies are the most effective approaches for drowning detection …

A Uav Target Tracking And Control Algorithm Based On Siamrpn, Songming Jiao, Hui Ding, Yufei Zhong, Xin Yao, Jiahao Jiahao Zheng

Journal of System Simulation

Aiming at the requirement of autonomously tracking land moving targets of rotary-wing UAVs, an autonomous and stable UAV tracking and control system that can adapt to the common interference environments such as scale changes, occlusions, and attitude changes is constructed.The system extracts the imaging position of the target in airborne camera through the twin network based on deep learning, and obtains the relative pose of the target. The image processing algorithm is designed to process the icons in the tracking frame, and the yaw angle of UAV relative to the tracking target is obtained, Kalman filter is introduced to …

Data-Driven Deep Learning-Based Analysis On Thz Imaging, Haoyan Liu

Graduate Theses and Dissertations

Breast cancer affects about 12.5% of women population in the United States. Surgical operations are often needed post diagnosis. Breast conserving surgery can help remove malignant tumors while maximizing the remaining healthy tissues. Due to lacking effective real-time tumor analysis tools and a unified operation standard, re-excision rate could be higher than 30% among breast conserving surgery patients. This results in significant physical, physiological, and financial burdens to those patients. This work designs deep learning-based segmentation algorithms that detect tissue type in excised tissues using pulsed THz technology. This work evaluates the algorithms for tissue type classification task among freshly …

Research On Visual Inspection Algorithm Of Crimping Appearance Defects For Wiring Harness Terminals, Bingan Yuan, Mingen Zhong, Jingxin Ni

Journal of System Simulation

Abstract: Aiming at the low efficiency and high missing rate of wiring harness terminals, an image detection method based on machine vision is proposed. The characteristic parameters of five typical defects in three main parts of wiring harness terminals are analyzed and defined. Tthe algorithms of extracting positioning datum, segmenting inspected-parts adaptively, extracting the defect features and calculating the characteristic parameters are designed respectively, and the defects criterions are given. The experimental results show that the algorithms are suitable for single defect and multi-class defects, both the miss detection rate and the false positiveness rate are low. The accuracy and …

Image Provenance Analysis, Daniel Moreira, William Theisen, Walter Scheirer, Aparna Bharati, Joel Brogan, Anderson Rocha

Computer Science: Faculty Publications and Other Works

The literature of multimedia forensics is mainly dedicated to the analysis of single assets (such as sole image or video files), aiming at individually assessing their authenticity. Different from this, image provenance analysis is devoted to the joint examination of multiple assets, intending to ascertain their history of edits, by evaluating pairwise relationships. Each relationship, thus, expresses the probability of one asset giving rise to the other, through either global or local operations, such as data compression, resizing, color-space modifications, content blurring, and content splicing. The principled combination of these relationships unveils the provenance of the assets, also constituting an …

Tooth Position Determination By Automatic Cutting And Marking Of Dental Panoramic X-Ray Film In Medical Image Processing, Yen-Cheng Huang, Chiung-An Chen, Tsung-Yi Chen, He-Sheng Chou, Wei-Chi Lin, Tzu-Chien Li, Jia-Jun Yuan, Szu-Yin Lin, Chun-Wei Li, Shih-Lun Chen, Yi-Cheng Mao, Patricia Angela R. Abu, Wei-Yuan Chiang, Wen-Shen Lo

Department of Information Systems & Computer Science Faculty Publications

This paper presents a novel method for automatic segmentation of dental X-ray images into single tooth sections and for placing every segmented tooth onto a precise corresponding position table. Moreover, the proposed method automatically determines the tooth’s position in a panoramic X-ray film. The image-processing step incorporates a variety of image-enhancement techniques, including sharpening, histogram equalization, and flat-field correction. Moreover, image processing was implemented iteratively to achieve higher pixel value contrast between the teeth and cavity. The next image-enhancement step is aimed at detecting the teeth cavity and involves determining the segment and points separating the upper and lower jaw, …

A Two-Stage Hair Region Localization Method For Guided Laser Hair Removal, Murat Avşar, İmam Şami̇l Yeti̇k

Turkish Journal of Electrical Engineering and Computer Sciences

Removal of hair using laser is a widely used method, where our goal is to permanently remove hair by using laser to cause heat in order to thermally damage the hair follicle. However, currently available laser hair removal systems affect the outer skin layers besides hair follicles. This is a disadvantage of classical methods with major health risks. We propose a method to overcome these health risks by guiding the laser beam only to automatically localized hair regions. This study aims to develop an automated feature-based hair region localization method as an integral part of the proposed hair removal system …

A Novel Hybrid Decision-Based Filter And Universal Edge-Based Logical Smoothingadd-On To Remove Impulsive Noise, Rajanbir Singh Ghumaan, Prateek Jeet Singh Sohi, Nikhil Sharma, Bharat Garg

Turkish Journal of Electrical Engineering and Computer Sciences

This paper presents a novel hybrid filter along with a universal extension to remove salt and pepper noise even at a very high noise density. The proposed filter initially specifies a threshold and then denoises the image using a combination of linear, nonlinear, and probabilistic techniques. Furthermore, to improve the quality, a universal add-on is presented which uses edge detection and smoothening techniques to brush out fine details from the restored image. To evaluate the efficacy, the proposed and existing filtering techniques are implemented in MATLAB and simulated with benchmark images. The simulation results show that the proposed filter is …

Application Of Artificial Intelligence And Geographic Information System For Developing Automated Walkability Score, Md Mehedi Hasan

Dissertations

Walking is considered as one of the major modes of active transportation, which contributes to the livability of cities. It is highly important to ensure walk friendly sidewalks to promote human physical activities along roads. Over the last two decades, different walk scores were estimated in respect to walkability measures by applying different methods and approaches. However, in the era of big data and machine learning revolution, there is still a gap to measure the composite walkability score in an automated way by applying and quantifying the activityfriendliness of walkable streets. In this study, a street-level automated walkability score was …

Deep Learning For Remote Sensing Image Processing, Yan Lu

Computational Modeling & Simulation Engineering Theses & Dissertations

Remote sensing images have many applications such as ground object detection, environmental change monitoring, urban growth monitoring and natural disaster damage assessment. As of 2019, there were roughly 700 satellites listing “earth observation” as their primary application. Both spatial and temporal resolutions of satellite images have improved consistently in recent years and provided opportunities in resolving fine details on the Earth's surface. In the past decade, deep learning techniques have revolutionized many applications in the field of computer vision but have not fully been explored in remote sensing image processing. In this dissertation, several state-of-the-art deep learning models have been …

Measurement Error Correction Model Of Tof Depth Camera, Le Wang, Luo Yu, Haikuan Wang, Minru Fei

Journal of System Simulation

Abstract: 3D data intuitively reflects the full view of the target or scene. Time of Flight (ToF) camera is a range imaging sensor that can provide 3D geometric information of targets immediately, thus it is widely applied in the robot positioning and navigation, 3D reconstruction and other aspects etc. Due to the operational principle of the camera itself, there are variety measurement errors of the source data obtained by ToF, resulting in image distortion. The measurement errors in the imaging process of ToF camera were analyzed and summarized, and the cubic spline interpolation method combined with look-up table was proposed …

Vehicle Logo Recognition Based On Sparse Sampling And Gradient Distribution Features, Binbin Zhou, Shangbing Gao, Zhigeng Pan, Liangliang Wang, Hongyang Wang

Journal of System Simulation

Abstract: The vehicle logo location and recognition are separated in the traditional method, the location errors will affect the subsequent recognition, at the same time the vehicle logo images are with low resolution and poor quality. Thus, a novel method was proposed which integrated the vehicle logo location and recognition organically. The sample images were sampled by sparse sampling, and then the point set was divided into adjacent point set and non adjacent point set, and the gradient feature and light and dark feature were extracted respectively, constructing the feature library. The logo coarse location area was multi-scale scanned. The …

Image Denoising Using Deep Convolutional Autoencoder With Feature Pyramids, Ekrem Çeti̇nkaya, Mustafa Furkan Kiraç

Turkish Journal of Electrical Engineering and Computer Sciences

Image denoising is 1 of the fundamental problems in the image processing field since it is the preliminary step for many computer vision applications. Various approaches have been used for image denoising throughout the years from spatial filtering to model-based approaches. Having outperformed all traditional methods, neural-network-based discriminative methods have gained popularity in recent years. However, most of these methods still struggle to achieve flexibility against various noise levels and types. In this paper, a deep convolutional autoencoder combined with a variant of feature pyramid network is proposed for image denoising. Simulated data generated by Blender software along with corrupted …

Identifying Regional Trends In Avatar Customization, Peter Mawhorter, Sercan Sengun, Haewoon Kwak, D. Fox Harrell

Research Collection School Of Computing and Information Systems

Since virtual identities such as social media profiles and avatars have become a common venue for self-expression, it has become important to consider the ways in which existing systems embed the values of their designers. In order to design virtual identity systems that reflect the needs and preferences of diverse users, understanding how the virtual identity construction differs between groups is important. This paper presents a new methodology that leverages deep learning and differential clustering for comparative analysis of profile images, with a case study of almost 100 000 avatars from a large online community using a popular avatar creation …

Low-Energy Acceleration Of Binarized Convolutional Neural Networks Using A Spin Hall Effect Based Logic-In-Memory Architecture, Ashkan Samiee, Payal Borulkar, Ronald F. Demara, Peiyi Zhao, Yu Bai

Engineering Faculty Articles and Research

Deep Learning (DL) offers the advantages of high accuracy performance at tasks such as image recognition, learning of complex intelligent behaviors, and large-scale information retrieval problems such as intelligent web search. To attain the benefits of DL, the high computational and energy-consumption demands imposed by the underlying processing, interconnect, and memory devices on which software-based DL executes can benefit substantially from innovative hardware implementations. Logic-in-Memory (LIM) architectures offer potential approaches to attaining such throughput goals within area and energy constraints starting with the lowest layers of the hardware stack. In this paper, we develop a Spintronic Logic-in-Memory (S-LIM) XNOR neural …

Seeing Eye To Eye: A Machine Learning Approach To Automated Saccade Analysis, Maigh Attre

Honors Scholar Theses

Abnormal ocular motility is a common manifestation of many underlying pathologies particularly those that are neurological. Dynamics of saccades, when the eye rapidly changes its point of fixation, have been characterized for many neurological disorders including concussions, traumatic brain injuries (TBI), and Parkinson’s disease. However, widespread saccade analysis for diagnostic and research purposes requires the recognition of certain eye movement parameters. Key information such as velocity and duration must be determined from data based on a wide set of patients’ characteristics that may range in eye shapes and iris, hair and skin pigmentation [36]. Previous work on saccade analysis has …

Image Processing Algorithms For Elastin Lamellae Inside Cardiovascular Arteries, Mahmoud Habibnezhad

Department of Computer Science and Engineering: Dissertations, Theses, and Student Research

Automated image processing methods are greatly needed to replace the tedious, manual histology analysis still performed by many physicians. This thesis focuses on pathological studies that express the essential role of elastin lamella in the resilience and elastic properties of the arterial blood vessels. Due to the stochastic nature of the shape and distribution of the elastin layers, their morphological features appear as the best candidates to develop a mathematical formulation for the resistance behavior of elastic tissues. However, even for trained physicians and their assistants, the current measurement procedures are highly error-prone and prolonged. This thesis successfully integrates such …

Elimination Of Useless Images From Raw Camera-Trap Data, Ulaş Tekeli̇, Yalin Baştanlar

Turkish Journal of Electrical Engineering and Computer Sciences

Camera-traps are motion triggered cameras that are used to observe animals in nature. The number of images collected from camera-traps has increased significantly with the widening use of camera-traps thanks to advances in digital technology. A great workload is required for wild-life researchers to group and label these images. We propose a system to decrease the amount of time spent by the researchers by eliminating useless images from raw camera-trap data. These images are too bright, too dark, blurred, or they contain no animals. To eliminate bright, dark, and blurred images we employ techniques based on image histograms and fast …

Recognition Of Incomplete Objects Based On Synthesis Of Views Using A Geometric Based Local-Global Graphs, Michael Christopher Robbeloth

Browse all Theses and Dissertations

The recognition of single objects is an old research field with many techniques and robust results. The probabilistic recognition of incomplete objects, however, remains an active field with challenging issues associated to shadows, illumination and other visual characteristics. With object incompleteness, we mean missing parts of a known object and not low-resolution images of that object. The employment of various single machine-learning methodologies for accurate classification of the incomplete objects did not provide a robust answer to the challenging problem. In this dissertation, we present a suite of high-level, model-based computer vision techniques encompassing both geometric and machine learning approaches …

Accurate And Compact Stochastic Computations By Exploiting Correlation, Hamdan Abdellatef, Mohamed Khalil Hani, Nasir Shaikh-Husin

Turkish Journal of Electrical Engineering and Computer Sciences

Recent studies have shown, contrary to what was previously believed, that by exploiting correlation in stochastic computing (SC) designs, more accurate SC circuits with low area cost can be realized. However, if these basic SC circuits or blocks are cascaded in series to form a large complex system, correlation between stochastic numbers (SNs) from one block to the next would be lost; thus, inaccuracies are introduced. In this study, we propose correlating circuits to be used in building complex correlated SC systems. One of the circuits is the correlator that restores lost correlations between two SNs due to previous processing. …

A Multiseed-Based Svm Classification Technique For Training Sample Reduction, Imran Sharif, Debasis Chaudhuri

Turkish Journal of Electrical Engineering and Computer Sciences

A support vector machine (SVM) is not a popular method for a very large dataset classification because the training and testing time for such data are computationally expensive. Many researchers try to reduce the training time of SVMs by applying sample reduction methods. Many methods reduced the training samples by using a clustering technique. To reduce its high computational complexity, several data reduction methods were proposed in previous studies. However, such methods are not effective to extract informative patterns. This paper demonstrates a new supervised classification method, multiseed-based SVM (MSB-SVM), which is particularly intended to deal with very large datasets …

Image Stitching And Matching Tool In The Automated Iterative Reverse Engineer (Aire) Integrated Circuit Analysis Suite, David C. Bowman

Browse all Theses and Dissertations

Due to current market forces, leading-edge semiconductor fabrication plants have moved outside of the US. While this is not a problem at first glance, when it comes to security-sensitive applications, over-production, device cloning, or design alteration becomes a possibility. Since these vulnerabilities exist during the fabrication phase, a Reverse Engineering (RE) step must be introduced to help ensure secure device operation. This thesis proposes several unique methods and a collection of tools to ensure trust assurance in integrated circuit design by detecting fabrication flaws and possible hardware Trojans using several image processing techniques; fused into a singular view of the …

Analysis Of 3d Cone-Beam Ct Image Reconstruction Performance On A Fpga, Devin Held

Electronic Thesis and Dissertation Repository

Efficient and accurate tomographic image reconstruction has been an intensive topic of research due to the increasing everyday usage in areas such as radiology, biology, and materials science. Computed tomography (CT) scans are used to analyze internal structures through capture of x-ray images. Cone-beam CT scans project a cone-shaped x-ray to capture 2D image data from a single focal point, rotating around the object. CT scans are prone to multiple artifacts, including motion blur, streaks, and pixel irregularities, therefore must be run through image reconstruction software to reduce visual artifacts. The most common algorithm used is the Feldkamp, Davis, and …

Video Annotation By Crowd Workers With Privacy-Preserving Local Disclosure, Apeksha Dipak Kumavat

Open Access Theses

Advancements in computer vision are still not reliable enough for detecting video content including humans and their actions. Microtask crowdsourcing on task markets such as Amazon Mechnical Turk and Upwork can bring humans into the loop. However, engaging crowd workers to annotate non-public video footage risks revealing the identities of people in the video who may have a right to anonymity.

This thesis demonstrates how we can engage untrusted crowd workers to detect behaviors and objects, while robustly concealing the identities of all faces. We developed a web-based system that presents obfuscated videos to crowd workers, and provides them with …

Some Properties Of Digital H-Spaces, Özgür Ege, İsmet Karaca

Turkish Journal of Electrical Engineering and Computer Sciences

In this paper, we study certain properties of digital H-spaces. We prove that a digital image that has the same digital homotopy type with any digital H-space is also a digital H-space. We show that the digital fundamental group of a digital H-space is abelian. We give examples that are related to a digital homotopy associative H-space and a $\kappa $-contractible digital H-space. Several important applications of digital H-spaces are given in computer vision and image processing. Finally, we deal with the importance of digital H-space in digital topology and image processing. We conclude that any $\kappa $-contractible digital image …

Brain Tumor Detection Using Monomodal Intensity Based Medical Image Registration And Matlab, Emrah Irmak, Ergun Erçelebi̇, Ahmet Hani̇fi̇ Ertaş

Turkish Journal of Electrical Engineering and Computer Sciences

digital image processing. Using suitable computer programming techniques and transformation between two images, a new much more informative image can be found. In this paper, three important and basic medical image registration (MIR) methods, namely MIR by maximization of mutual information, MIR using cross correlation (Fourier transform approach), and MIR by minimization of similarity metric, were proposed and accordingly two comprehensive applications were performed using MIR by minimization of the similarity metric, which uses the sum of the squared differences metric as a metric and the regular step gradient descent optimizer as an optimizer. What is more, MR images of …

Hardware Implementation Of A Scale And Rotation Invariant Object Detection Algorithm On Fpga For Real-Time Applications, Murat Peker, Hali̇s Altun, Fuat Karakaya

Turkish Journal of Electrical Engineering and Computer Sciences

A hardware implementation of a computationally light, scale, and rotation invariant method for shape detection on FPGA is devised. The method is based on histogram of oriented gradients (HOG) and average magnitude difference function (AMDF). AMDF is used as a decision module that measures the similarity/dissimilarity between HOG vectors of an image in order to classify the object. In addition, a simulation environment implemented on MATLAB is developed in order to overcome the time-consuming and tedious process of hardware verification on the FPGA platform. The simulation environment provides specific tools to quickly implement the proposed methods. It is shown that …

Gpu/Cpu Performance Of Image Processing Tasks For Use In The Cam 2 System, Jonathan Cottom, Yung-Hsiang Lu, Young-Sol Koh

The Summer Undergraduate Research Fellowship (SURF) Symposium

Over the past several years, graphics processing units (GPU) have increasingly been viewed as the future of image processing engines. Currently, the Continuous Analysis of Many CAMeras (CAM²) project performs its processing on CPUs, which will potentially be more costly as the system scales to service more users. This study seeks to analyze the performance gains of GPU processing and evaluate the advantage of supporting GPU-accelerated analysis for CAM² users. The platform for comparing the CPU and GPU performance has been the NVIDIA Jetson TK1. The target hardware implementation is an Amazon cloud instance, where final cost …

Design And Implementation Of Digital Information Security For Physical Documents, Pengcheng Wang

Masters Theses

The objective of this thesis is to improve the security for physical paper documents. Providing information security has been difficult in environments that rely on physical paper documents to implement business processes. Our work presents the design of a digital information security system for paper documents, called "CryptoPaper", that uses 2-dimensional codes to represent data and its security properties on paper. A special scanner system is designed for "CryptoPaper" which uses image recognition techniques and cloud-based access control to display plaintext of encrypted and encoded data to authorized users.