Aydin Ayanzadeh: publications and news https://www.ayanzadeh.com/ New papers and research updates from Aydin Ayanzadeh, UMBC. en Sat, 14 Nov 2026 00:00:00 +0000 Modular Deep Learning Mechanisms for Auditable Next-Day Wildfire Spread Prediction https://www.ayanzadeh.com/nextday-wildfire/ https://www.ayanzadeh.com/nextday-wildfire/ Preprint arXiv 2026, 2026. Modular deep learning augmentations for next-day wildfire spread prediction: wind- and slope-conditioned attention, retrieval-augmented correction, and dual-stream gating. Characterizing an On-Device Depth-Aware Multimodal Workload in Assistive Navigation for Blind and Low-Vision Users https://www.ayanzadeh.com/iiswc-2026-workload/ https://www.ayanzadeh.com/iiswc-2026-workload/ Conference paper IISWC 2026, 2026. An end-to-end characterization of an on-device, depth-aware multimodal workload for assistive indoor navigation for blind and low-vision users. LLM-Guided Agentic Floor Plan Parsing for Accessible Indoor Navigation of Blind and Low-Vision People https://www.ayanzadeh.com/cvpr-2026-floorplan-agentic/ https://www.ayanzadeh.com/cvpr-2026-floorplan-agentic/ Conference paper CVPR 2026 Workshop (GRAIL-V), 2026. An agentic framework that converts a single floor plan image into a structured, retrievable knowledge base for safe, accessible indoor navigation. WildfireVLM: AI-Powered Analysis for Early Wildfire Detection Using Satellite Imagery https://www.ayanzadeh.com/wildfirevlm-page/ https://www.ayanzadeh.com/wildfirevlm-page/ Conference paper IGARSS 2026, 2026. A vision-language model for early wildfire detection and risk assessment from satellite imagery. Floorplan2Guide: LLM-Guided Floorplan Parsing for BLV Indoor Navigation https://www.ayanzadeh.com/IEEEBigData2025/ https://www.ayanzadeh.com/IEEEBigData2025/ Conference paper IEEE BigData 2025, 2025. Floorplan2Guide transforms floorplans into spatial knowledge graphs and language-guided indoor navigation for blind and low-vision users. PURSUhInT: In Search of Informative Hint Points Based on Layer Clustering for Knowledge Distillation https://www.ayanzadeh.com/pursuhint/ https://www.ayanzadeh.com/pursuhint/ Journal article Expert Systems with Applications, 2023. A clustering-based hint selection method for knowledge distillation: teacher layers are clustered and the cluster centers become the hint points. Automated analysis of phase-contrast optical microscopy time-lapse images: application to wound healing and cell motility assays of breast cancer https://www.ayanzadeh.com/wound-healing-microscopy-chapter/ https://www.ayanzadeh.com/wound-healing-microscopy-chapter/ Book chapter Book chapter, Academic Press, 2023. An end-to-end workflow for analyzing phase-contrast microscopy data from cell motility and wound healing assays. Improved cell segmentation using deep learning in label-free optical microscopy images https://www.ayanzadeh.com/cell-segmentation-microscopy/ https://www.ayanzadeh.com/cell-segmentation-microscopy/ Journal article Turkish Journal of Electrical Engineering and Computer Sciences, 2021. A modified U-Net with a pretrained ResNet-18 encoder, residual decoder blocks, and a redesigned skip-connection for cell segmentation in label-free microscopy images. Representation learning using graph autoencoders with residual connections https://www.ayanzadeh.com/resvgae-graph-autoencoders/ https://www.ayanzadeh.com/resvgae-graph-autoencoders/ Preprint arXiv 2021, 2021. Residual connections for deep graph autoencoders: ResVGAE, a deep variational graph autoencoder with multiple residual modules, for link prediction. Deep Learning based Segmentation Pipeline for Label-Free Phase-Contrast Microscopy Images https://www.ayanzadeh.com/siu-2020-segmentation-pipeline/ https://www.ayanzadeh.com/siu-2020-segmentation-pipeline/ Conference paper SIU 2020, 2020. A deep learning segmentation pipeline for label-free phase-contrast microscopy images, improving U-Net and FPN with an ImageNet-pretrained ResNet-18 encoder. Graph Embedding For Link Prediction Using Residual Variational Graph Autoencoders https://www.ayanzadeh.com/siu-2020-graph-embedding/ https://www.ayanzadeh.com/siu-2020-graph-embedding/ Conference paper SIU 2020, 2020. Residual Variational Graph Autoencoder (RVGAE): residual connections boost variational graph autoencoders for link prediction. A New Class of Scaling Matrices for Scaled Trust Region Algorithms https://www.ayanzadeh.com/scaling-matrices-trust-region/ https://www.ayanzadeh.com/scaling-matrices-trust-region/ Preprint arXiv 2019, 2019. A new class of affine scaling matrices for interior point Newton-type methods, defined as a convex combination of well-known scaling matrices. Cell Segmentation of 2D Phase-Contrast Microscopy Images with Deep Learning Method https://www.ayanzadeh.com/tiptekno-2019-cell-segmentation/ https://www.ayanzadeh.com/tiptekno-2019-cell-segmentation/ Conference paper TIPTEKNO 2019, 2019. A multiresolution U-Net with multi-combination augmentation for cell segmentation in 2D phase-contrast microscopy images. Automated Segmentation of Cells in Phase Contrast Optical Microscopy Time Series Images https://www.ayanzadeh.com/tiptekno-2019-cell-time-series/ https://www.ayanzadeh.com/tiptekno-2019-cell-time-series/ Conference paper TIPTEKNO 2019, 2019. Traditional image processing and deep learning approaches for automated cell segmentation from phase-contrast microscopy time series, evaluated against manually annotated datasets. Modified Deep Neural Networks for Dog Breeds Identification https://www.ayanzadeh.com/dog-breeds-identification/ https://www.ayanzadeh.com/dog-breeds-identification/ Preprint Preprints 2018, 2018. Fine-tuned pretrained ImageNet models for dog breed identification, reaching 89.66% test accuracy on the Stanford Dogs dataset. A Modified Ant colony Based Approach to Digital Image Edge Detection https://www.ayanzadeh.com/ant-colony-edge-detection/ https://www.ayanzadeh.com/ant-colony-edge-detection/ Conference paper KBEI 2015, 2015. A modified ant colony optimization method for digital image edge detection, faster and more efficient than former ant colony based methods. Gaussian Three-Dimensional kernel SVM for Edge Detection Applications https://www.ayanzadeh.com/gaussian-svm-edge-detection/ https://www.ayanzadeh.com/gaussian-svm-edge-detection/ Conference paper Conference 2015 ยท arXiv 2017, 2015. A support vector machine with a three-dimensional Gaussian radial basis function kernel for edge detection, outperforming Canny and Sobel detectors. M.S. Graduation & International Alumni Scholarship https://www.ayanzadeh.com/#news news-2025-05-15-m-s-graduation-international-alumni-scholarship Achievement Thu, 15 May 2025 00:00:00 +0000 Presenting at CMD-IT/ACM Richard Tapia Conference https://www.ayanzadeh.com/#news news-2025-01-presenting-at-cmd-it-acm-richard-tapia-conference Conference New Paper: Improved Autoencoder for Nuclei Segmentation https://www.ayanzadeh.com/#news news-2024-12-15-new-paper-improved-autoencoder-for-nuclei-segmentation Research Sun, 15 Dec 2024 00:00:00 +0000 Presented at STARS Celebration Conference https://www.ayanzadeh.com/#news news-2024-10-presented-at-stars-celebration-conference Conference WildfireVLM: Environmental AI Research https://www.ayanzadeh.com/projects/wildfire-vlm.html news-2024-wildfirevlm-environmental-ai-research Project PURSUhInT Published in Expert Systems with Applications https://www.ayanzadeh.com/pursuhint/ news-2023-pursuhint-published-in-expert-systems-with-applications Publication Vodafone FutureLab Research Fellowship https://www.ayanzadeh.com/#news news-2019-05-vodafone-futurelab-research-fellowship Award