Results 71 to 80 of about 8,226 (198)

Feedback‐Driven Medical Image Segmentation: Modelling on Boundary‐Sensitive Feature Preservation

CAAI Transactions on Intelligence Technology, EarlyView.
ABSTRACT Accurate segmentation of medical images, particularly along ambiguous lesion boundaries, plays a vital role in clinical diagnosis and treatment planning. Conventional segmentation approaches often struggle to recover fine‐grained contours due to spatial detail loss caused by convolutional downsampling and the inherent sparsity of edge features,
Jikai Yang, Ming Yang, Zhiming Li, Shenglong Chen   +3 more
wiley   +1 more source

Classifying Deepfakes Using Swin Transformers

CoRR
3 ...
Aprille J. Xi, Eason Chen
openaire   +2 more sources

Frequency Domain and Gradient-Spatial Multi-Scale Swin KANsformer for Remote Sensing Scene Classification

Remote Sensing
Transformer-based deep learning techniques have recently shown outstanding potential in remote sensing scene classification (RSSC), benefiting from their ability to capture global semantic relationships and contextual dependencies.
Xiaozhang Zhu, Junqing Huang, Haihui Wang   +2 more
doaj   +1 more source

A Wildlife Multi‐Modal Recognition Method Based on Infrared Camera Images Using MWAC

Integrative Zoology, EarlyView.
A multimodal wild wildlife classification model MWAC consists of two parts:(1) Wavelet KAN module (WDS‐KAN) for image feature extraction, and (2) Label Correlated Text Extraction module (LC‐BERT) for text feature extraction. ABSTRACT Nowadays, infrared camera technology has become an important tool for large‐scale monitoring and assessment of wildlife ...
Siming Deng, Guoxiong Zhou, Genhua Liu, Zhishu Xiao, Xiaoqun Huang, Yang Hu, Ling Gui   +6 more
wiley   +1 more source

A Lightweight Deep Neural Network for Individual Leopard Identification: Supporting Movement Ecology in the Taihang Mountains of Henan

Integrative Zoology, EarlyView.
We developed LeoNet, a lightweight deep learning model for leopard individual identification. It achieved 97.42% accuracy and was robust to age‐related pattern changes and unfamiliar individuals. Applied to camera‐trap data from the Taihang Mountains, LeoNet identified 41 wild North China leopards and generated individual‐level records to support ...
Jing He, Guoqing Shangguan, Yu Li, Fuming Ma, Zhishu Xiao, Mingliang Fan, Zhuo Chen, Junzhan Ren, Yingfeng Ren, Yunlong Wang, Tianping Wang, Hua Yang, Hualei Shen, Xiaohong Chen   +13 more
wiley   +1 more source

Visualizing Image Segmentation Network Behavior Through the Lens of Scale Space Analysis

Computer Graphics Forum, EarlyView.
Abstract Deep neural networks are widely used for image segmentation, also in sensitive applications such as medical imaging or autonomous driving. However, few explainable AI methods are available that help developers understand such networks beyond classification.
A. C. Mikliss, T. Schultz
wiley   +1 more source

Multi-scale capsule Swin Transformer-based method for SAR image target recognition

Tongxin xuebao
A multi-scale capsule Swin Transformer network (MSCSTN) was proposed by synergizing the semantic feature encoding of capsule units with the context feature mapping of Swin Transformer. Capsule encoding and the Swin Transformer were jointly applied to SAR
HOU Yuchao, WANG Jie, LI Hongtao, HAO Yan, DUAN Xiaoqi, HUANG Kaiwen, TIAN Youliang   +6 more
doaj  

Comparing deep learning models for butterfly and moth (Lepidoptera) species identification

Insect Conservation and Diversity, EarlyView.
Deep learning models open new possibilities for the processing of image‐based species records. We used high‐performance computing to compare 40 models for butterfly and moth species identification based on a citizen science dataset with over 500,000 images of 162 species.
Friederike Barkmann, Andreas Lindner, Johannes Rüdisser   +2 more
wiley   +1 more source

Deep Learning Models for Detection of Periapical Radiolucent Lesions on Panoramic Radiographs: A Systematic Review and Meta‐Analysis

International Endodontic Journal, EarlyView.
ABSTRACT Background Panoramic radiographs are used routinely to screen dental conditions and treatment patterns. Recently, numerous studies have suggested that deep learning (DL) models can be utilized for analysing panoramic radiographs. Objective This review aimed to evaluate the accuracy of DL models in detecting periapical radiolucent lesions (PRLs)
Ibrahim Ali Ahmad, Raidan Ba‐Hattab, Sadeq Ali Al‐Maweri, Faleh Tamimi   +3 more
wiley   +1 more source

FML-Swin: An Improved Swin Transformer Segmentor for Remote Sensing Images

IEEE Access
Semantic segmentation of urban remote sensing images is a very challenging task. Due to the complex background, occlusion overlap and small scale target of urban remote sensing image, the semantic segmentation results have some defects such as target confusion and similarity, target boundary ambiguity, and small scale target omission.
Tianren Wu, Wenqin Deng, Rui Lin, Junzhe Jiang 0004, Xueyun Chen   +4 more
openaire   +2 more sources