A Lightweight and Efficient Plant Disease Detection Method Integrating Knowledge Distillation and Dual-Scale Weighted Convolutions by Xiong Yang, Hao Wang, Qi Zhou, Lei Lu, Lijuan Zhang, Changming Sun, Guilu Wu

“…Subsequently, we developed the DSConv module—a novel convolutional structure employing double-scale weighted convolutions that dynamically adjust to different scale perceptions and optimize attention allocation. This module replaces the conventional Conv module in YOLOv10. …”
Get full text

Uncovering the Effects of Different Formulae of Milk Powders on the Fecal Microorganisms and Metabolites of Bengal Tiger (<i>Panthera tigris</i> spp. <i>tigris</i>) Cubs by Xuanzhen He, Tingting Xiao, Jing Fang, Peng Zhang, Shenghui Luo, Sufang Han, Yuansheng Wu, Lizhen Li, Zhihao Cao, Yuhan Ji, Guixin Dong, Baichuan Deng

“…In order to optimize diets for Bengal tiger cubs and improve their health condition and survival rates, we conducted microbiota and metabolomics analyses on fecal samples from Bengal tiger cubs fed goat and dog milk replacer formulae. …”
Get full text

Prediction of the Characteristics of Concrete Containing Crushed Brick Aggregate by Marijana Hadzima-Nyarko, Miljan Kovačević, Ivanka Netinger Grubeša, Silva Lozančić

“…The study focuses on optimizing key parameters like the minimum leaf size in the models. …”
Get full text

Research on foreign object intrusion detection in railway tracks based on MSL-YOLO by Hongxia Niu, Dingchao Feng, Tao Hou

“…To address these issues, this paper proposes MSL-YOLO, a lightweight and accurate object detector optimized for railway applications. Specifically, a Multi-scale Shared Convolution Module (MSCM) is designed to replace SPPF, enhancing feature extraction while reducing parameters and computational cost. …”
Get full text

Using Graph Neural Networks in Reinforcement Learning With Application to Monte Carlo Simulations in Power System Reliability Analysis by Oystein Rognes Solheim, Boye Annfelt Hoverstad, Magnus Korpas

“…Recent efforts from the authors indicate that optimal power flow solvers could potentially be replaced with the policies of deep reinforcement learning agents, to obtain significant speedups of Monte Carlo simulations while retaining close to optimal accuracies. …”
Get full text

An In Situ Characterisation Method for 3-D Electrospun Foams by Kyriakos Almpanidis, Chloe J. Howard, Vlad Stolojan

“…Finally, this fast in situ evaluation method can replace the post-fabrication time-consuming one. It can be used as a fundamental step for an intelligent artificial intelligence tool that predicts optimal foam formation.…”
Get full text

Incremental Cotton Diseases Detection Based on Aerial Remote Sensing Image and Blockchain Sharding by Jing Nie, Haochen Li, Yang Li, Jingbin Li, Xuewei Chao, Sezai Ercisli

“…Optimize the neck network using deeply separable convolution to reduce parameters and computation. …”
Get full text

Addressing Deficits With Crowdfunded Journalism by Kenneth Gibson

“…Crowdfunding’s ability to help replace the previous financial paradigm depends on understanding how best to make direct financial solicitations to news audiences. …”
Get full text

Effects of Colemanite and Fiber Types on Strength and Water Absorption of Mortar Mixtures Containing Air-Entraining Admixture by Ayşe İrem Ozansoy, Süleyman Özen

“…This study not only contributes to reducing the carbon footprint of cementitious systems by partially replacing cement with colemanite but also provides practical insights to optimize the use of fibers to increase strength performance and reduce water absorption properties from durability parameters. …”
Get full text

Subsea Nodule Recognition and Deployment Detection Method Based on Improved YOLOv8s by Jixin Li, Junchao Li, Bin Su, Yuxin Cui

“…Additionally, SK-Conv convolution modules replace certain Conv and C2f layers in the backbone network, while the Focal SIoU loss function replaces the original CIoU loss function. …”
Get full text

Characterization of the oxygen-tolerant formate dehydrogenase from Clostridium carboxidivorans by Eva-Maria Brouwer, Hitesh K. R. Medipally, Hitesh K. R. Medipally, Saskia Schwab, Shanshan Song, Marc M. Nowaczyk, Marc M. Nowaczyk, Marc M. Nowaczyk, Martin Hagemann, Martin Hagemann

“…During the production of recombinant ccFDH, in which the selenocysteine codon was replaced by Cys, we were able to replace the W with Mo as the transition metal in the ccFDH metal cofactor, resulting in a two-fold increase of 6 μmol formate min−1 in enzyme activity. …”
Get full text

Lightweight insulator target detection algorithm based on improved YOLOX by Bing Zeng, Wei Hua, Dezhi Li, Zhihao Zhou, Hao Wan, Yunmin Xie, Tangbing Li, Yucong Chen, Jianglei Li, Shenli Wang, Shixun Fu, Zihan Jin, Wenhua Zhang

“…Finally, the σWIoU loss function replaces the traditional IoU to accelerate convergence and optimize overall performance. …”
Get full text

Enhanced Efficiency of CZTS Solar Cells with Reduced Graphene Oxide and Titanium Dioxide Layers: A SCAPS Simulation Study by Dounia Fatihi, Giorgio Tseberlidis, Vanira Trifiletti, Simona Binetti, Eleonora Isotta, Paolo Scardi, Abderrafi Kamal, R’hma Adhiri, Narges Ataollahi

“…Another option is to replace the cadmium sulfide (CdS) buffer layer with a titanium dioxide (TiO₂) layer. …”
Get full text

Impact of Alternative Feed Ingredients and Feeding Strategies on Growth, Muscle Morphology, and Fillet Quality of Genetically Selected Gilthead Seabream (<i>Sparus aurata</i>) in a... by Paula Sarmiento, Pedro L. Castro, Rafael Ginés

“…The 85AS feeding strategy optimized fillet quality by preventing lipid accumulation and muscle adaptations, as observed with the other feeding strategies. …”
Get full text

DGS-Yolov7-Tiny: a lightweight pest and disease target detection model suitable for edge computing environments by Ping Yu, Baoshu Zong, Xiaozhong Geng, Hui Yan, Baijin Liu, Cheng Chen, Hupeng Liu, Xiaoqing Xu

“…A novel fusion convolution, DGSConv, replaced the standard convolutions and effectively reduced the number of parameters while retaining detailed feature information. …”
Get full text

YOLOGX: an improved forest fire detection algorithm based on YOLOv8 by Caixiong Li, Yue Du, Xing Zhang, Peng Wu

“…Finally, the proposed Focal-SIoU loss function replaces the original loss function, effectively reducing directional errors by combining angle, distance, shape, and IoU losses, thus optimizing the model training process. …”
Get full text

Research on injection molded parts defect detection algorithm based on multiplicative feature fusion and improved attention mechanism by Rongnan Zhang, Yang Li, Zhiguang Guan

“…This method introduces several key improvements: (1) the MAFHead, a four-detection head based on multiplicative feature fusion, which replaces the original detection head to enhance feature representation; (2) the RepGFPN-SE module, a re-parameterized generalized feature pyramid network that improves detection of small objects by replacing the original C2f. module; (3) and the BiNorma module, employing a bi-level routing attention mechanism to optimize the training process by reducing input distribution changes across layers. …”
Get full text

Maintaining Efficient Operation of Low-Yield Wells During the Development of the Cenomanian Deposit by E. I. Inyakina, E. E. Levitina, A. V. Plazun, M. J. Z. Alsheikhly

“…For efficient operation of wells, it is necessary to create optimal technological operating modes for water removal and carry out measures to limit water inflows. …”
Get full text

Influence of the effect of recycled fine aggregate and powder on the frost resistance of recycled concrete by Quan Ma, Wei Yang, Min Song, Zhenhua Duan

“…To verify the feasibility of replacing cementitious materials with recycled fine powder (RFP), cement paste was prepared by substituting RFP for cementitious materials by mass fraction at 0, 10%, 20%, and 30%. …”
Get full text

GCS-YOLO: A Lightweight Detection Algorithm for Grape Leaf Diseases Based on Improved YOLOv8 by Qiang Hu, Yunhua Zhang

“…The lightweight feature extraction module C2f-GR is proposed to replace the C2f module. C2f-GR achieves lightweight design while effectively capturing detailed features of multi-scale information by replacing partial convolutions in C2f with Ghost Modules. …”
Get full text