Article
Agriculture, Multidisciplinary
Lei Shen, Shan Chen, Zhiwen Mi, Jinya Su, Rong Huang, Yuyang Song, Yulin Fang, Baofeng Su
Summary: This study proposes a method combining deep learning and image analysis to accurately identify veraison in colored wine grapes under natural field growing conditions. The study utilizes semantic segmentation model for background removal and constructs a Mask R-CNN instance segmentation pipeline with optimized anchor parameters. The ResNet50-FPN backbone network performs the best in Mask R-CNN. Additionally, a method for characterizing berry veraison using the H component of HSV color space is proposed and an algorithm is developed to identify veraison progress based on the percentage of berries in different grades. The proposed method achieves high test accuracy and can provide important reference for automated monitoring and intelligent management decisions of grape growth.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Environmental Sciences
Mengying Cao, Ying Sun, Xin Jiang, Ziming Li, Qinchuan Xin
Summary: The study utilized deep learning to predict the leaf phenology of deciduous broadleaf forests at 56 sites in North America using PhenoCam images. In the one-site experiment, high accuracy was achieved, but lower accuracies were observed in the all-site experiment. The model's accuracy improved when deep networks used region of interest images as inputs instead of entire images.
Review
Computer Science, Information Systems
Ednawati Rainarli, Suprapto, Wahyono
Summary: The rapid development of scene text detection highlights the need for text recognition in scene images. This review analyzes the related research of scene text detection in the last decade, discussing the strengths and weaknesses of different methods and showcasing the application of deep learning in text detection. Researchers have been focusing on detecting horizontal text, multi-orientation text, multilingual text, curved text, and arbitrary-shaped text.
COMPUTER SCIENCE REVIEW
(2021)
Article
Business
Wenjun Tu, Xiaolan Zheng, Lei Li, Zhiang (John) Lin
Summary: Chinese firms' increasing cross-border acquisitions challenge existing theories of multinational enterprise due to unique characteristics and government involvement. Our study shows that firms with more government ownership have better post-CBA performance, but this relationship is influenced by firm-level and country-level boundary conditions.
INTERNATIONAL BUSINESS REVIEW
(2021)
Article
Agronomy
Alec Zuo, Sarah Ann Wheeler, Ying Xu
Summary: Australia, specifically the southern Murray-Darling Basin, has the most advanced water market in the world. This study identified five clusters of irrigators in the region, including those trying to expand the farm, diversify, downsize, transition, and save water. The findings suggest a trend of irrigation farm exit, particularly among older farmers facing financial and/or psychological stress, due to the predicted rise in temperature caused by climate change.
AGRICULTURAL WATER MANAGEMENT
(2022)
Article
Biology
Yang Liu, Xiaoyun Zhong, Shiyao Zhai, Zhicheng Du, Zhenyuan Gao, Qiming Huang, Can Yang Zhang, Bin Jiang, Vijay Kumar Pandey, Sanyang Han, Runming Wang, Yuxing Han, Chuhui Wang, Peiwu Qin
Summary: This article proposes a method to improve CPR instruction qualification using action segmentation and deep learning techniques, and experimental results on a CPR dataset demonstrate its effectiveness.
COMPUTERS IN BIOLOGY AND MEDICINE
(2023)
Article
Statistics & Probability
Maxime Rischard, Zach Branson, Luke Miratrix, Luke Bornn
Summary: This study examines the premium on house price for a particular school district in New York City using a novel implementation of a geographic regression discontinuity design (GeoRDD). By modeling spatial structure with Gaussian processes regression, the research identifies significant price differences along borders, with one border showing a statistically significant 20% higher price for houses on the more desirable side.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2021)
Article
Environmental Sciences
Ruijing Zheng, Mengqi Qiu, Yaping Wang, Deyang Zhang, Zeping Wang, Yu Cheng
Summary: With the acceleration of urbanization, waste classification has become an inevitable factor threatening the environment and human health. Through a questionnaire survey, we found that the elderly, women, and people with higher education are more likely to participate in waste classification. Attitude, collaborative governance, and institutional pressure positively affect waste classification behavior, while subjective norm and infrastructure have a negative effect.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Article
Economics
Charles Shaaba Saba, Oladipo Olalekan David
Summary: This study examines the convergence of telecommunication infrastructures in 205 countries from 2000 to 2018. The findings suggest that there is panel convergence for the full sample, developed, less developed, upper-middle and high-income countries, while low- and lower-middle-income countries diverge. The System GMM results further suggest that economic, human capital, financial deepening and demographic factors play a significant positive role in telecommunication infrastructure convergence across countries.
JOURNAL OF THE KNOWLEDGE ECONOMY
(2023)
Article
Multidisciplinary Sciences
Shayan Mostafaei, Minh Tuan Hoang, Pol Grau Jurado, Hong Xu, Lluis Zacarias-Pons, Maria Eriksdotter, Saikat Chatterjee, Sara Garcia-Ptacek
Summary: This study used machine learning algorithms to identify important variables associated with mortality risk in dementia patients. They found that age, sex, body mass index, mini-mental state examination score, time periods before and after diagnosis were significant predictors of mortality risk. The study also discovered new variables not previously reported to be associated with mortality in dementia. These findings demonstrate the potential of machine learning algorithms in predicting mortality risk in clinical settings.
SCIENTIFIC REPORTS
(2023)
Article
Construction & Building Technology
Xiaofei Yang, Enrique del Rey Castillo, Yang Zou, Liam Wotherspoon
Summary: Deep learning techniques are used to automate the classification of bridge point clouds, and two synthetic data augmentation strategies are proposed to address data scarcity and erroneous boundary segmentation issues. Evaluation experiments show that the synthetic data augmentation strategies can significantly improve the training sample scarcity problem.
AUTOMATION IN CONSTRUCTION
(2023)
Article
Engineering, Electrical & Electronic
Lihui Wang, Qiucheng Shen
Summary: A boundary-aware semantic segmentation algorithm is proposed to improve the precision and robustness of welding zone inspection system, achieving a segmentation accuracy of 93.70% and an average detection time of less than 11 ms per image. The algorithm effectively reduces interference from non-weld zones by guiding boundary information in the semantic segmentation process.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Engineering, Electrical & Electronic
Riktim Mondal, Showmik Bhowmik, Ram Sarkar
Summary: The segmentation of touching components in handwritten documents is a crucial task for document image processing. A generative adversarial network (GAN)-based model named tsegGAN is developed in this work to address this issue, showing significant performance improvement compared to state-of-the-art GAN models.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2021)
Article
Engineering, Environmental
Zhongyao Liang, Yaoyang Xu, Gang Zhao, Wentao Lu, Zhenghui Fu, Shuhang Wang, Tyler Wagner
Summary: This study proposes a novel framework using quantile regression and interpretable machine learning to identify factors forcing ecological observations to approach the upper boundary and explores its implications for ecosystem management.
FRONTIERS OF ENVIRONMENTAL SCIENCE & ENGINEERING
(2023)
Article
Physics, Fluids & Plasmas
Clement Scherding, Georgios Rigas, Denis Sipp, Peter J. Schmid, Taraneh Sayadi
Summary: In this paper, a model-agnostic machine-learning technique is proposed to extract a reduced thermochemical model of a gas mixture from a library, reducing the cost of simulating hypersonic flows. By clustering and generating surrogate surfaces, different thermochemical states are handled, and the method is validated to improve solver performance by up to 70% while maintaining accuracy.
PHYSICAL REVIEW FLUIDS
(2023)
Article
Computer Science, Artificial Intelligence
Chenglong Li, Xiaobin Yang, Guohao Wang, Aihua Zheng, Chang Tan, Jin Tang
Summary: License plate recognition is crucial in various practical applications, however, recognizing license plates of large vehicles is challenging due to low resolution, contamination, low illumination, and occlusion. To address this problem, a novel data generation framework based on the Disentangled Generation Network is proposed to ensure the generation diversity and integrity for robust enlarged license plate recognition.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Sara Casao, Alvaro Serra-Gomez, Ana C. Murillo, Wendelin Bohmer, Javier Alonso-Mora, Eduardo Montijano
Summary: This paper presents a hybrid camera system that combines static and mobile cameras, exploiting the cooperation between tracking and control modules to achieve high-level scene understanding. The static camera network provides global awareness, while the mobile cameras enhance the information about the people on the scene.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Anh-Dzung Doan, Bach Long Nguyen, Surabhi Gupta, Ian Reid, Markus Wagner, Tat-Jun Chin
Summary: To ensure reliable object detection in autonomous systems, the detector needs to adapt to changes in appearance caused by environmental factors. We propose a selective adaptation approach using domain gap as a criterion to improve the efficiency of the detector's operation.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Tianhong Dai, Wei Li, Xilei Cao, Jianzhuang Liu, Xu Jia, Ales Leonardis, Youliang Yan, Shanxin Yuan
Summary: This study proposes a novel frequency-guided deep neural network (FHDRNet) for high dynamic range (HDR) imaging from multiple low dynamic range (LDR) images, aiming to address ghosting artifacts. By conducting HDR fusion in the frequency domain, the network utilizes low-frequency signals to remove specific ghosting artifacts and high-frequency signals to preserve details. Extensive experiments demonstrate that this approach achieves state-of-the-art performance.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Guobin Li, Reyer Zwiggelaar
Summary: Breast cancer is the most commonly diagnosed female malignancy worldwide. Recent developments in deep convolutional neural networks have shown promising performance for breast cancer detection and classification. However, biased features can be learned due to variations in appearance and small datasets. To address this issue, a densely connected convolutional network (DenseNet) was trained using texture features representing different physical morphological representations as inputs. The use of connectivity estimation and nearest neighbors improved the network's unbiased prediction. The approach achieved higher diagnostic accuracy and provided visual explanations for model predictions.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Yuezun Li, Cong Zhang, Honggang Qi, Siwei Lyu
Summary: Deep Neural Networks (DNNs) are vulnerable to adversarial perturbations, limiting their applicability in safe-critical scenarios. To address this, a new method called AdaNI is proposed to increase feature randomness through adaptive noise injection, improving adversarial robustness. Extensive experiments demonstrate the efficacy of AdaNI against various white-box and black-box attacks, as well as its applicability in DeepFake detection.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Chengyin Hu, Weiwen Shi, Ling Tian, Wen Li
Summary: In this study, we introduce a pioneering black-box light-based physical attack called Adversarial Neon Beam (AdvNB). Our method excels in attack modeling, efficient attack simulation, and robust optimization, striking a balance between robustness and efficiency. Through rigorous evaluation, we achieve impressive attack success rates in both digital and real-world scenarios. AdvNB demonstrates its stealthiness through comparisons with baseline samples and consistently achieves high success rates when targeting advanced DNN models.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Hang Wang, Zhenyu Ding, Cheng Cheng, Yuhai Li, Hongbin Sun
Summary: Learning-based super resolution has made remarkable progress in improving image quality, but the performance decreases when the degradation kernel changes. Blind SR networks can estimate the degradation kernel and adapt well in realistic scenarios, improving performance and runtime. This paper proposes a design that imposes constraints for the kernel estimation network in both the image domain and kernel domain, resulting in high-quality images and efficient runtime.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Yuantao Chen, Runlong Xia, Kai Yang, Ke Zou
Summary: This paper proposes an improved image inpainting network using a multi-scale feature module and improved attention module. The network addresses issues in deep learning-based image inpainting algorithms, such as information loss in deep level features and the neglect of semantic features. The proposed network generates better inpainting results by reducing information loss and enhancing the ability to restore texture and semantic features.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
Article
Computer Science, Artificial Intelligence
Yi-Tung Chan
Summary: This study proposes a novel maritime background subtraction method based on ensemble learning theory to address the challenges posed by dynamic marine environments and noise, improving the detection accuracy and enhancing maritime transportation security for autonomous ships in open waters.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2024)
