Article
Engineering, Civil
Thaweesak Trongtirakul, Sos Agaian, Adel Oulefki, Karen Panetta
Summary: Several oil spill disasters in the past decade have posed a major threat to the marine ecosystem, damaging marine life and causing economic losses. Developing a cost-effective oil spill detection system that includes source identification, extent estimation, transport path analysis, and weather conditions is crucial. Thermal and polarimetric imagery are promising sensing modalities that can enhance oil spill detection compared to conventional imaging techniques. This article investigates the usage of thermal and polarimetric cameras for tracking 3D oil spills in the sea and proposes robust unsupervised oil sensing algorithms.
IEEE JOURNAL OF OCEANIC ENGINEERING
(2023)
Article
Geochemistry & Geophysics
Tao Zhang, Wei Wang, Zhen Yang, Junjun Yin, Jian Yang
Summary: This study investigates the relationship between the polarimetric covariance matrix and complete polarimetric covariance difference matrix, and proposes a novel ship detection method SPAN(SDP) that can detect small ships more accurately. Experimental results show that the new method is more effective than other state-of-the-art methods.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2021)
Article
Remote Sensing
Alfonso Lopez, Juan M. Jurado, Carlos J. Ogayar, Francisco R. Feito
Summary: This paper introduces a framework that successfully connects different types of aerial images through multi-layer registration using a registration method that is invariant to intensity differences. Correction of deformed images as the first step of the registration algorithm lays the foundation for more advanced systems.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2021)
Article
Physics, Multidisciplinary
Xiaolin Xu, Yuan Zong, Cheng Lu, Xingxun Jiang
Summary: This paper introduces a method called Enhanced Sample Self-Revised Network (ESSRN) to address the issue of outlier samples in cross-dataset facial expression recognition. Experimental results demonstrate that ESSRN can effectively reduce the negative impact of outlier samples on cross-dataset FER.
Article
Multidisciplinary Sciences
Martin Gjoreski, Ivana Kiprijanovska, Simon Stankoski, Ifigeneia Mavridou, M. John Broulidakis, Hristijan Gjoreski, Charles Nduka
Summary: This study used a novel wearable surface electromyography to investigate the affective states induced by different videos. The results showed that subjective valence, subjective arousal, and objective valence measured through sEMG varied significantly depending on the video content.
SCIENTIFIC REPORTS
(2022)
Article
Computer Science, Artificial Intelligence
Yanli Ji, Yuhan Hu, Yang Yang, Heng Tao Shen
Summary: This paper proposes a Region Attention eNhanced Domain Adaptation (RANDA) approach for unsupervised cross-domain facial expression recognition. It uses an iterative pseudo label assignment method to generate pseudo labels in the target domain and employs adversarial learning to confuse feature representation of facial expressions in the source and target domains. In addition, a facial landmark guided fine-grained region attention learning module is designed to enhance significant emotion features and weaken domain discrepancy. Experimental results show that RANDA outperforms state-of-the-art approaches in multiple datasets, providing an effective solution for cross-domain FER.
IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Tao Yang, Yuhang Zhang, Jie Sun, Xun Wang
Summary: Homomorphic encryption is a key method for protecting user privacy in cloud computing, especially for face recognition systems. By balancing workloads and re-implementing similarity measurement functions, the homomorphic encryption version achieves similar performance to the original version.
NEURAL PROCESSING LETTERS
(2022)
Article
Chemistry, Analytical
Mustafa Al Qudah, Ahmad Mohamed, Syaheerah Lutfi
Summary: This paper proposes three classification models to address occlusion challenges in thermal facial images, specifically eyeglasses and facial hair. These models are able to classify six basic spontaneous emotions, and the results obtained were promising and comparable to those of other studies.
Article
Engineering, Electrical & Electronic
Sevket Demirci
Summary: In this study, the capabilities of polarimetric target decomposition in SAR-based ATR applications are investigated using 2D turntable ISAR imagery of a T-72 tank. The results of different decomposition methods are analyzed and compared in terms of identifying the physical scattering mechanisms. The usefulness of various secondary parameters is also discussed. The results demonstrate that the decomposition features can be effectively utilized in subsequent ATR processes.
INTERNATIONAL JOURNAL OF ANTENNAS AND PROPAGATION
(2022)
Article
Computer Science, Information Systems
Ziyang Zhang, Xiang Tian, Yuan Zhang, Kailing Guo, Xiangmin Xu
Summary: This study proposes enhanced discriminative global-local feature learning with priority (EDGL-FLP), which focuses on feature extraction without auxiliary information and feature fusion based on priority. EDGL-FLP achieves state-of-the-art performance on the FER benchmarks RAF-DB, SFEW, AffectNet, FED-RO, and MMI with accuracies of 89.63%, 62.31%, 61.09%, 71.42%, and 86.47%, respectively. Thus, EDGL-FLP is robust for both in-the-wild and in-the-lab FER datasets.
INFORMATION SCIENCES
(2023)
Article
Optics
Yidong Luo, Junchao Zhang, Di Tian
Summary: This paper proposes a demosaicking model based on sparse coding to reconstruct full-resolution images from color polarization mosaic images. The model considers RGB-polarization channels correlation, adaptive sub-dictionaries, and non-local self-similarity restrictions. Experimental results show that the proposed method outperforms the current state-of-the-art method in terms of quantitative measures and visual quality.
OPTICS AND LASERS IN ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Ercheng Pei, Zhanxuan Hu, Lang He, Hailong Ning, Abel Diaz Berenguer
Summary: This paper proposes an ensemble learning-enhanced multitask network architecture and a novel adaptive weighted loss-based multitask learning strategy for continuous affect recognition. Experimental results demonstrate the potential of the proposed method compared to state-of-the-art methods.
EXPERT SYSTEMS WITH APPLICATIONS
(2024)
Article
Computer Science, Information Systems
Meng Xia, Zhijie Wang, Fang Han, Yanting Kang
Summary: This paper proposes an augmented multi-dimensional and multi-grained Cascade Forest for cloud/snow recognition, which has good recognition efficiency by capturing spatial and spectral information of cloud/snow satellite imagery, and introduces a simple augmentation method to enhance the robustness of cloud/snow recognition.
Article
Environmental Sciences
Junjun Yin, Jian Yang
Summary: A unified reconstruction framework for the general CP mode is proposed in the study, which is based on formalized CP descriptors and a three-component decomposition method. Additionally, a least squares estimation method is extended to arbitrary CP modes to solve the system of non-linear equations.
Article
Engineering, Biomedical
Weijun Gong, Yurong Qian, Weihang Zhou, Hongyong Leng
Summary: The recognition of dynamic facial expressions is challenging due to various factors, and obtaining discriminative expression features has been difficult. Traditional deep learning networks lack understanding of global and temporal expressions. This study proposes an enhanced spatial-temporal learning network to improve dynamic facial expression recognition.
BIOMEDICAL SIGNAL PROCESSING AND CONTROL
(2024)
Article
Optics
Alex J. Yuffa, Yael Gutierrez, Juan M. Sanz, Rodrigo Alcaraz de la Osa, Jose M. Saiz, Francisco Gonzalez, Fernando Moreno, Gorden Videen
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION
(2015)
Article
Optics
Miroslav Kocifaj, Frantisek Kundracik, Gorden Videen, Alex J. Yuffa, Jozef Klacka
JOURNAL OF QUANTITATIVE SPECTROSCOPY & RADIATIVE TRANSFER
(2016)
Article
Optics
Alex J. Yuffa, Yael Gutierrez, Juan M. Sanz, Rodrigo Alcaraz de la Osa, Jose M. Saiz, Francisco Gonzalez, Fernando Moreno, Gorden Videen
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION
(2016)
Article
Optics
Johannes Markkanen, Alex J. Yuffa
JOURNAL OF QUANTITATIVE SPECTROSCOPY & RADIATIVE TRANSFER
(2017)
Article
Engineering, Electrical & Electronic
Alex J. Yuffa, Johannes Markkanen
IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION
(2018)
Article
Optics
Qiang Sun, Evert Klaseboer, Alex J. Yuffa, Derek Y. C. Chan
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION
(2020)
Article
Optics
Qiang Sun, Evert Klaseboer, Alex J. Yuffa, Derek Y. C. Chan
JOURNAL OF THE OPTICAL SOCIETY OF AMERICA A-OPTICS IMAGE SCIENCE AND VISION
(2020)
Article
Acoustics
Marc Andrew Valdez, Alex J. Yuffa, Michael B. Wakin
Summary: We propose a compressive sampling method for reconstructing acoustic fields based on field measurements on a pre-defined spherical grid. This method establishes the relations between signal sparsity, measurement number, and reconstruction accuracy. In comparison to traditional methods, the proposed method uses equiangular grid sampling and transforms the reconstruction problem into a multi-dimensional Fourier domain problem. Experimental results show that this method outperforms classical Nyquist sampling and requires fewer measurements.
JOURNAL OF THE ACOUSTICAL SOCIETY OF AMERICA
(2022)
Article
Engineering, Electrical & Electronic
Marc Andrew Valdez, Alex J. Yuffa, Michael B. Wakin
Summary: In this paper, we prove a compressive sensing guarantee on the rotation group by defining Slepian functions on a measurement sub-domain and transforming the inverse problem to the Slepian functions. By requiring measurements on a select-able sub-domain, our approach provides higher accuracy and reduces the number of measurements compared to other methods using Wigner D-functions. Numerical examples demonstrate the superiority of our method in reconstruction quality.
IEEE TRANSACTIONS ON SIGNAL PROCESSING
(2022)
Article
Physics, Multidisciplinary
J. Alex Yuffa
Summary: Green's theorem and Green's identities are fundamental concepts that have applications across various branches of science and mathematics. This paper discusses a vector analogue of Green's three scalar identities and examines their uses, while also providing historical context related to the work of George Green.
JOURNAL OF PHYSICS COMMUNICATIONS
(2021)
Proceedings Paper
Engineering, Electrical & Electronic
Johannes Markkanen, Alex J. Yuffa, Joshua A. Gordon
2018 INTERNATIONAL APPLIED COMPUTATIONAL ELECTROMAGNETICS SOCIETY SYMPOSIUM (ACES)
(2018)
Proceedings Paper
Engineering, Electrical & Electronic
Alexandra E. Curtin, David R. Novotny, Alex J. Yuffa, Selena Leitner
2017 ANTENNA MEASUREMENT TECHNIQUES ASSOCIATION SYMPOSIUM (AMTA)
(2017)
Article
Optics
Alex J. Yuffa, Vadym Kaydash, Viktor Korokhin, Yuriy Shkuratov, Evgenij Zubko, Gorden Videen
Article
Optics
Yong-Le Pan, Chuji Wang, Leonid A. Beresnev, Alex J. Yuffa, Gorden Videen, David Ligon, Joshua L. Santarpia
Article
Optics
Nathaniel J. Short, Alex J. Yuffa, Gorden Videen, Shuowen Hu