Dirk Colbry

Dirk Colbry

Auto-generated profile, based on publicly available information

Dirk Colbry

Michigan State University

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
3 July 2023 in Journal of Cleaner Production
Verified
Journal Article
Closed access
MCMeng Cai
Meng Cai
EKEva Kassens-Noor
Eva Kassens-Noor
ZZZhiqiang Zhao
Zhiqiang Zhao
DCDirk Colbry
Dirk Colbry

26 July 2020
Verified
Conference Paper
Open access
HNHenry Neeman
Henry Neeman
DADavid Akin
David Akin
HAHussein Al-Azzawi
Hussein Al-Azzawi
KBKevin L. Brandt
Kevin L. Brandt
JKJamene Brooks Kieffer
Jamene Brooks Kieffer
DVDaniel Voss
Daniel Voss
Cyberinfrastructure (CI) Facilitation is the process of helping researchers to use research computing systems and services to advance their computing/data-intensive research goals. The growing need for CI Facilitation isn’t being met by traditional academic degree/certificate programs, so informal education is required. Since 2015, the Virtual Residency Program (VRP) has been teaching key CI Facilitation skills to pre-service and in-service CI Facilitators, at introductory, intermediate and advanced levels, via a combination of (a) workshops, (b) conference calls, and (c) apprenticeships. The demand for CI Facilitation has expanded exponentially, with the known CI Facilitator population growing from dozens in the late 2000s to over a thousand presently, so the need for this kind of training has become increasingly acute. The 2019 VRP workshop was presented at a combined introductory/intermediate level, with content suitable for both new and experienced CI Facilitators, to attract both of these populations, in many cases presenting the same topic first at the introductory level and then intermediate, to maximize impact. Since 2015, the VRP has served 924 participants from 370 institutions in every US state and 3 US territories plus 11 other countries.

8 April 2024 in Journal of Open Source Software
Verified
Journal Article
Open access
PDF file
MCMeng Cai
Meng Cai
Yingjie LiYingjie Li
Yingjie Li
DCDirk Colbry
Dirk Colbry
VFVeronica F. Frans
Veronica F. Frans
YZYuqian Zhang
Yuqian Zhang
Seesus is an open-source Python package that evaluates whether a textual expression aligns with the concept of sustainability as defined by the United Nations Sustainable Development Goals (SDGs). It labels a statement with the 17 SDGs as well as 169 specific targets and categorizes the statement into social, environmental, or economic sustainability. Compared to existing SDG classification software in Python, seesus is the only one that is tuned to a finer scale-the SDG target level-and offers transparency and controllability, allowing users to examine and customize the syntax of text classification according to their needs.

6 September 2024 in ELCVIA Electronic Letters on Computer Vision and Image Analysis
Verified
Journal Article
Open access
PDF file
MBMaryam Berijanian
Maryam Berijanian
KGKatrina Gensterblum
Katrina Gensterblum
DMDoruk Alp Mutlu
Doruk Alp Mutlu
KRKatelyn Reagan
Katelyn Reagan
AHAndrew Hart
Andrew Hart
DCDirk Colbry
Dirk Colbry
This work introduces two new distance metrics for comparing labeled arrays, which are common outputs of image segmentation algorithms. Each pixel in an image is assigned a label, with binary segmentation providing only two labels ('foreground' and 'background'). These can be represented by a simple binary matrix and compared using pixel differences. However, many segmentation algorithms output multiple regions in a labeled array. We propose two distance metrics, named LAD and MADLAD, that calculate the distance between two labeled images. By doing so, the accuracy of different image segmentation algorithms can be evaluated by measuring their outputs against a 'ground truth' labeling. Both proposed metrics, operating with a complexity of O(N) for images with N pixels, are designed to quickly identify similar labeled arrays, even when different labeling methods are used. Comparisons are made between images labeled manually and those labeled by segmentation algorithms. This evaluation is crucial when searching through a space of segmentation algorithms and their hyperparameters via a genetic algorithm to identify the optimal solution for automated segmentation, which is the goal in our lab, SEE-Insight. By measuring the distance from the ground truth, these metrics help determine which algorithm provides the most accurate segmentation.

22 September 2026 in ELCVIA Electronic Letters on Computer Vision and Image Analysis
Verified
Journal Article
Open access
MBMaryam Berijanian
Maryam Berijanian
AKAbdi Kusata
Abdi Kusata
MFMax Fahrenkrug
Max Fahrenkrug
ASAmin Sehati
Amin Sehati
DCDirk Colbry
Dirk Colbry
This research explores the potential of state-of-the-art Generative Adversarial Networks (GANs) for segmentation through image-to-image (I2I) translation. The presented model, seGANmentation, combines convolutional neural networks and vision transformers to generate segmentation labels directly from input images. It aims to reduce manual labeling and enhance segmentation performance using synthetic data. Comparative analysis with U-Net as a benchmark in segmentation, shows U-Net achieving an average of 0.99 in accuracy, Dice Coefficient, and Intersection over Union, while seGANmentation scores 0.90, 0.83, and 0.71, respectively. Although it did not outperform U-Net, it highlights the potential for GAN-based segmentation. This work lays the foundation for employing GANs in this novel approach. This paper's proposed framework could be valuable in data-scarce scenarios, enhance task accuracy with synthetic data augmentation, reduce the need for extensive training pairs by refining GAN-generated labels with off-the-shelf segmentors, and achieve unsupervised segmentation, unlike traditional neural networks. These hypotheses remain to be tested in future studies.