A5057736319- Advanced Data Compression Techniques
- Image and Signal Denoising Methods
- Algorithms and Data Compression
- Career Development and Diversity
- Image Enhancement Techniques
- Video Coding and Compression Technologies
- Advanced Image and Video Retrieval Techniques
- Higher Education Research Studies
- Digital Filter Design and Implementation
- Image and Video Quality Assessment
- Hand Gesture Recognition Systems
- COVID-19 and Mental Health
- Mental Health Research Topics
- Advanced Vision and Imaging
- Image Retrieval and Classification Techniques
- Mentoring and Academic Development
- Engineering Education and Curriculum Development
- Gender Diversity and Inequality
- Speech and Audio Processing
- Medical Imaging Techniques and Applications
- Engineering Education and Pedagogy
- Indoor and Outdoor Localization Technologies
- Tactile and Sensory Interactions
- Digital Mental Health Interventions
- Medical Image Segmentation Techniques
Stevens Institute of Technology
2022-2024
University of Washington
2012-2024
California Polytechnic State University
2023-2024
Washington Center
2024
California State Polytechnic University
2023
Seattle University
2006-2015
Washington College
2015
Cal Poly Corporation
2015
Google (United States)
2011
Massachusetts Institute of Technology
2005
We present the unequal loss protection (ULP) framework in which amounts of forward error correction are applied to progressive data provide graceful degradation image quality as packet losses increase. develop a simple algorithm that can find good assignment within ULP framework. use set partitioning hierarchical trees coder this work, but our protect any compression scheme. In addition, we promote PMF expected channel conditions so system work with almost model or estimate losses. when...
P.H. Westerink et al. (1988) developed an optimal bit allocation algorithm that is simplified when all operational distortion-rate functions (which are referred to as quantizer functions) convex. This restated using the generalized Breiman, Friedman, Olshen, and Stone (BFOS) (a recently technique for variable rate vector design) both cases of convex nonconvex (QFs), its complexity analyzed. The use BFOS analysed. It shown if each source has a function then low.< <ETX...
There is a growing body of research revealing that longitudinal passive sensing data from smartphones and wearable devices can capture daily behavior signals for human modeling, such as depression detection. Most prior studies build evaluate machine learning models using collected single population. However, to ensure model work larger group users, its generalizability needs be verified on multiple datasets different populations. We present the first evaluating cross-dataset models,...
It is well known that women are significantly underrepresented in scientific fields the United States, and computer science no exception. As of 1987- 1988, constituted slightly more than half U.S. population 45% employed workers U.S., but they made up only 30% scientists. Moreover, 10% doctoral-level During same time period, 20% physicians and, at doctoral level, 35% psychologists, 22% life scientists, mathematicians On other hand, there some disciplines which represent an even smaller...
A technique for directly designing a variable-rate tree-structured vector quantizer by growing the tree one node at time rather than layer is presented. The natural extension of method decision trees. When pruned with generalized algorithm optimally pruning trees, improvement measured in signal-to-noise ratio high rates over fixed-rate same initial rate. can be interpreted as constrained inverse algorithm.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML"...
The prevalence of mobile phones and wearable devices enables the passive capturing modeling human behavior at an unprecedented resolution scale. Past research has demonstrated capability sensing to model aspects physical health, mental education, work performance, etc. However, most algorithms models proposed in previous follow a one-size-fits-all (i.e., population modeling) approach that looks for common behaviors amongst all users, disregarding fact individuals can behave very differently,...
A review is presented of vector quantization, the mapping pixel intensity vectors into binary indexing a limited number possible reproductions, which popular image compression algorithm. Compression has traditionally been done with little regard for processing operations that may precede or follow step. Recent work used quantization both to simplify tasks, such as enhancement classification, halftoning, and edge detection, reduce computational complexity by performing tasks simultaneously...
The impact of COVID-19 on students has been enormous, with an increase in worries about fiscal and physical health, a rapid shift to online learning, increased isolation. In addition these changes, disabilities/health concerns may face accessibility problems learning or communication tools, their stress be compounded by additional risks such as financial pre-existing conditions. To our knowledge, no one looked specifically at the concerns. this paper, we present data from survey 147 without...
This mixed-method study examined the experiences of college students during COVID-19 pandemic through surveys, experience sampling data collected over two academic quarters (Spring 2019 n 1 = 253; Spring 2020 2 147), and semi-structured interviews with 27 undergraduate students. There were no marked changes in mean levels depressive symptoms, anxiety, stress, or loneliness between 2020, course term. Students both cohort who indicated psychosocial vulnerability at initial assessment showed...
We present an algorithm that assigns unequal amounts of forward error correction to progressive data so as provide graceful degradation packet losses increase. use the SPIHT coder compress images in this work, but our can protect any compression scheme. The also almost function a model loss conditions. find for exponential with mean 20% and total rate 0.2 bpp, good image quality be obtained, even when 40% transmitted packets are lost.
This paper describes an algorithm that achieves approximately optimal assignment of forward error correction to progressive data within the unequal loss protection framework. It first finds under convex hull and fractional bit allocation assumptions. then relaxes those constraints find approximates global optimum. The has a running time O(hNlogN) where h is number points on source's utility-cost curve N packets transmitted.
We present a simple and efficient scheme for using the Set Partitioning in Hierarchical Trees (SPIHT) image compression algorithm generalized multiple description framework. To combat packet loss, controlled amounts of redundancy are added to original data during process. Unequal loss protection is implemented by varying amount with importance data. The achieves graceful degradation quality presence increasing loss; high obtained even when over half descriptions lost.
To evaluate the effects of lossy image (noninvertible) compression on diagnostic accuracy thoracic computed tomographic images.Sixty images from patients with mediastinal adenopathy and pulmonary nodules were compressed to six different levels tree-structured vector quantization. Three radiologists then used original for diagnosis. Unlike many previous receiver operating characteristic-based studies that confidence rankings binary detection tasks, this study examined sensitivity predictive...
Three techniques for variable-rate vector quantizer design are applied to medical images. The first two extensions of an algorithm optimal pruning in tree-structured classification and regression due Breiman et al. code algorithms find subtrees a given (TSVQ), each one that it has the lowest average distortion all TSVQ with same or lesser rate. Since resulting have variable depth, natural coders result. third technique is joint optimization noiseless code. This relatively complex but...
Feeling a sense of belonging is central human motivation that has consequences for mental health and well-being, yet surprisingly little research examined how shapes among young adults. In three data sets from two universities (exploratory study: N = 157; Confirmatory Study 1: 121; 2: n 188 in winter term, 172 spring term), we found lower levels daily-assessed feelings early across the academic term predicted higher depressive symptoms at end term. Furthermore, these relationships held when...
The authors apply a lossy compression algorithm to medical images, and quantify the quality of images by diagnostic performance radiologists, as well traditional signal-to-noise ratios subjective ratings. authors' study is unlike previous studies effects in that they consider nonbinary detection tasks, simulate actual practice instead using paired tests or confidence rankings, use statistical methods are more appropriate for clinical data than popular receiver operating characteristic...
For Deaf people, access to the mobile telephone network in United States is currently limited text messaging, forcing communication English as opposed American Sign Language (ASL), preferred language. Because ASL a visual language, video phones have potential give people real-time their However, even today's best compression techniques can not yield intelligible at cell phone bandwidths. Motivated by this constraint, we conducted one focus group and user study with members of Community...
A "discount" version of Q-methodology for HCI, called "HCI-Q", can be used in iterative design cycles to explore, from the point view users and other stakeholders, what makes technologies personally significant. Initially, designers critically reflect on their own assumptions about how a may affect social individual behavior. Then, use these as stimuli elicit people's points view. This process critical self-reflection evaluation helps designer assess fit between its intended context use. To...
Abstract The NSF-funded Redshirt in Engineering Consortium was formed 2016 with the goal of enhancing ability academically talented but underprepared students coming from low-income backgrounds to successfully graduate engineering degrees. takes its name practice redshirting college athletics, idea providing an extra year and support help promising complete a bachelor's degree. builds on success three existing "academic redshirt" programs expands model new schools. Existing Institutions...
Ubiquitous computing, the seamless integration of sensing, analytics, and feedback into daily life envisioned by Weiser [12], has come closer to reality with broad adoption smartphones wearable devices. These devices, integral users' routines, passively collect massive amounts data on human behavior, offering unprecedented insights personal health well-being [7]. For example, passive sensing can continuously monitor subtle changes in behavior indicative depression or other shifts mental...
Displacement estimated interframe (DEI) coding, a coding scheme for 3-D medical image data sets such as X-ray computed tomography (CT) or magnetic resonance (MR) images, is presented. To take advantage of the correlation between contiguous slices, displacement-compensated difference based on previous encoded. The best fitting distribution functions discrete cosine transform (DCT) coefficients obtained from displacement compensated images are determined and used in allocating bits optimizing...
A deep understanding of how discrimination impacts psychological health and well-being students could allow us to better protect individuals at risk support those who encounter discrimination. While the link between diminished physical is well established, existing research largely focuses on chronic long-term outcomes. short-term behavioral correlates events help concretely quantify such experiences, which in turn policy intervention design. In this paper we specifically examine, for first...