Custom Network Quantization Method for Lightweight CNN Acceleration on FPGAs by Lingjie Yi, Xianzhong Xie, Yi Wan, Bo Jiang, Junfan Chen

“…The low-bit quantization can effectively reduce the deep neural network storage as well as the computation costs. Existing quantization methods have yielded unsatisfactory results when being applied to lightweight networks. …”
Get full text

Comparison of Deep and Machine Learning Approaches for Quebec Tree Species Classification Using a Combination of Multispectral and LiDAR Data by Omid Reisi Gahrouei, Jean-François Côté, Philippe Bournival, Philippe Giguère, Martin Béland

“…The results indicated that Dense Convolution Network (DenseNet) achieved the best overall accuracy of 78%, outperforming machine learning methods by 5%. …”
Get full text

EPI-DynFusion: enhancer-promoter interaction prediction model based on sequence features and dynamic fusion mechanisms by Ao Zhang, Jianhua Jia, Mingwei Sun, Xin Wei

“…Furthermore, we incorporate the Convolutional Block Attention Module (CBAM) to enhance the model’s ability to focus on informative regions. …”
Get full text

Plant Disease Detection Using an Innovative Swin-Axial Transformer by Ao Zhang, Wei Liu

“…By introducing the TokenEmbedder module, the number of tokens is reduced, and multi-scale deep convolution is used to efficiently extract image features, significantly lowering computational costs. …”
Get full text

A Smartphone-Based Non-Destructive Multimodal Deep Learning Approach Using pH-Sensitive Pitaya Peel Films for Real-Time Fish Freshness Detection by Yixuan Pan, Yujie Wang, Yuzhe Zhou, Jiacheng Zhou, Manxi Chen, Dongling Liu, Feier Li, Can Liu, Mingwan Zeng, Dongjing Jiang, Xiangyang Yuan, Hejun Wu

“…Based on the lightweight MobileNetV2 network, a Multi-scale Dilated Fusion Attention module (MDFA) was designed to enhance the robustness of color feature extraction. A Temporal Convolutional Network (TCN) was then used to model dynamic patterns in chemical indicators across spoilage stages, combined with a Context-Aware Gated Fusion (CAG-Fusion) mechanism to adaptively integrate image and chemical temporal features. …”
Get full text

ACT-FRCNN: Progress Towards Transformer-Based Object Detection by Sukana Zulfqar, Zenab Elgamal, Muhammad Azam Zia, Abdul Razzaq, Sami Ullah, Hussain Dawood

“…Which integrates ACT with a Faster Region-Based Convolution Neural Network (FRCNN) for a detection task head. …”
Get full text

OEM-HWNet: A Prior Knowledge-Guided Network for Pavement Interlayer Distress Detection Based on Computer Vision Using GPR by Congde Lu, Senguo Cao, Xiao Wang, Guanglai Jin, Siqi Wang, Wenlong Cai

“…Firstly, an object enhancement module based on prior knowledge was designed to locate the regions of interlayer distress and enhance their characteristics. Then, wavelet convolution was added to increase the receptive field of the network and enhance the ability to capture low-frequency information. …”
Get full text

FEVT-SAR: Multicategory Oriented SAR Ship Detection Based on Feature Enhancement Vision Transformer by Minding Fang, Yu Gu, Dongliang Peng

“…FEViT includes two innovative lightweight modules: localized feature interactive convolution block (LFICB) and dual-granularity attention transformer block (DGTB). …”
Get full text

Estimating and mapping tailings properties of the largest iron cluster in China for resource potential and reuse: A new perspective from interpretable CNN model and proposed spectr... by Haimei Lei, Nisha Bao, Moli Yu, Yue Cao

“…Visible-near infrared-shortwave infrared (VIS-NIR-SWIR; 350–2500 nm) spectroscopy offers a rapid, non-destructive, and cost-effective method for quantitatively analyzing tailings properties. …”
Get full text

Deep Learning-Based Intelligent Detection Algorithm for Surface Disease in Concrete Buildings by Jing Gu, Yijuan Pan, Jingjing Zhang

“…First, we enhanced attention on multi-scale disease features by implementing structural reparameterization with parallel small-kernel expansion convolution. Next, we reconstructed the relationship between localization and classification tasks in the detection head and implemented dynamic selection of interactive features using a feature extractor to improve the accuracy of classification and recognition. …”
Get full text

Aided Greenway Design Approach Based on Internet Big Data and AIGC Fine-Tuning Model by Yifan WU, Lu MENG, Liang LI

“…Image features are extracted using the pre-trained convolutional neural network model Inception ResNetV2 and image data is clustered by K-means clustering algorithm. …”
Get full text

Lightweight Apple Leaf Disease Detection Algorithm Based on Improved YOLOv8 by LUO Youlu, PAN Yonghao, XIA Shunxing, TAO Youzhi

“…SPD-Conv was introduced to replace the original convolutional layers to retain fine-grained information and reduce model parameters and computational costs, thereby improving the accuracy of disease detection. …”
Get full text

Untrained perceptual loss for image denoising of line-like structures in MR images. by Elisabeth Pfaehler, Daniel Pflugfelder, Hanno Scharr

“…The uPL network's initialization is not important (e.g. for MR root images SSIM differences of 0.01 occur across initializations, while network depth and pooling operations impact denoising performance slightly more (SSIM of 0.83 for 5 convolutional layers and kernel size 3 vs. 0.86 for 5 convolutional layers and kernel size 5 for the root dataset). …”
Get full text

Early detection of Citrus Huanglongbing by UAV remote sensing based on MGA-UNet by Naibo Ye, Wenyong Mai, Feng Qin, Sen Yuan, Bo Liu, Zaiyuan Li, Conghui Liu, Fanghao Wan, Wanqiang Qian, Zhongzhen Wu, Xi Qiao

“…This study leverages multispectral imagery acquired via unmanned aerial vehicles (UAVs) and deep convolutional neural networks. This study introduce a novel model, MGA-UNet, specifically designed for HLB recognition. …”
Get full text

DM_CorrMatch: a semi-supervised semantic segmentation framework for rapeseed flower coverage estimation using UAV imagery by Jie Li, Chengyong Zhu, Chenbo Yang, Quan Zheng, Binhui Wang, Jingmin Tu, Qian Zhang, Sheng Liu, Xinfa Wang, Jiangwei Qiao

“…In this study, we propose a cost-effective and high-throughput approach using a semi-supervised learning framework, DM_CorrMatch. …”
Get full text

Transformers for Neuroimage Segmentation: Scoping Review by Maya Iratni, Amira Abdullah, Mariam Aldhaheri, Omar Elharrouss, Alaa Abd-alrazaq, Zahiriddin Rustamov, Nazar Zaki, Rafat Damseh

“…Currently, hybrid convolutional neural network-transformer architectures achieve state-of-the-art performances on benchmark datasets over standalone models. …”
Get full text

Research on geomagnetic indoor high-precision positioning algorithm based on generative model by Shuai MA, Ke PEI, Huayan QI, Hang LI, Wen CAO, Hongmei WANG, Hailiang XIONG, Shiyin LI

“…Aiming at the current bottleneck of constructing a fine geomagnetic fingerprint library that required a lot of labor costs, two generative models called the conditional variational autoencoder and the conditional confrontational generative network were proposed, which could collect a small number of data samples for a given location, and generate pseudo-label fingerprints.At the same time, in order to solve the problem of low positioning accuracy of single-point geomagnetic fingerprints, a geomagnetic sequence positioning algorithm based on attention mechanism of convolutional neural network-gated recurrent unit was designed, which could effectively use the spatial and temporal characteristics of fingerprints to achieve precise positioning.In addition, a real-time, portable mobile terminal data collection and positioning system was also designed and built.The actual test shows that the proposed model can effectively construct the available geomagnetic fingerprint database, and the average error of the proposed algorithm can reach 0.16 m.…”
Get full text

Divide-and-conquer routing for learning heterogeneous individualized capsules. by Hailei Yuan, Qiang Ren

“…Capsule Networks (CapsNets) have demonstrated an enhanced ability to capture spatial relationships and preserve hierarchical feature representations compared to Convolutional Neural Networks (CNNs). However, the dynamic routing mechanism in CapsNets introduces substantial computational costs and limits scalability. …”
Get full text

Analysis of Time-Fractional Delay Partial Differential Equations Using a Local Radial Basis Function Method by Kamran, Kalsoom Athar, Zareen A. Khan, Salma Haque, Nabil Mlaiki

“…The aim of utilizing the Laplace transform is to handle the costly convolution integral associated with the Caputo derivative and to avoid the effects of time-stepping techniques on the stability and accuracy of the numerical solution. …”
Get full text

Self-Supervised Learning of End-to-End 3D LiDAR Odometry for Urban Scene Modeling by Shuting Chen, Zhiyong Wang, Chengxi Hong, Yanwen Sun, Hong Jia, Weiquan Liu

“…Our method is as follows: (1) we efficiently encode 3D point cloud structures using voxel-based sparse convolution, and (2) we model inherent alignment uncertainty via covariance matrices, enabling novel self-supervised loss based on uncertainty modeling. …”
Get full text