Results 41 to 50 of about 2,016 (190)

Research on multi-objective detection method for incomplete information in coal mine underground

Meitan kexue jishu
Underground target detection technology in coal mines is an indispensable component of constructing a smart mine, providing real-time monitoring and recognition capabilities.
Lin SUN, Sheng CHEN, Xulong YAO, Yanbo ZHANG, Zhigang TAO, Peng LIANG   +5 more
doaj   +1 more source

MSFFNet: Multiscale Feature Fusion Network for Small Target Detection in Remote Sensing Images

CAAI Transactions on Intelligence Technology, Volume 11, Issue 2, Page 592-609, April 2026.
ABSTRACT With the advancement of satellite remote sensing technology, object detection based on high‐resolution remote sensing imagery has emerged as a prominent research focus in the field of computer vision. Although numerous algorithms have been developed for remote sensing image object detection, they still suffer from challenges such as low ...
Hui Zong, Chenbin Wang, Hao Wang, Xin Cheng, Shangbing Gao, Yunyang Yan   +5 more
wiley   +1 more source

An anchor-free object detector based on soften optimized bi-directional FPN [PDF]

, 2022
We propose an anchor-free object detector that combines a weighted bi-directional Feature Pyramid Network (BiFPN) and Soft Anchor Point Detector to address the object detection problem in a pixel-wise paradigm.
Zhang, T, Jia, W, Jin, B
core   +1 more source

SW‐YOLO: An Optimized YOLOv8 Architecture With ConvNeXt for Real‐Time and Accurate Substation Wiring Inspection

Engineering Reports, Volume 8, Issue 3, March 2026.
The study uses affine transformation, contrast enhancement, and mosaic masking for image enhancement and introduces the Convolutional NeXt module in YOLOv8 based on masked self‐encoder and response normalization, along with improvements to the convolutional block attention module.
Yu Lei, Zhihao Liang, Jiayun Huang
wiley   +1 more source

Empanada MitoNet model files

, 2022
PyTorch jit scripted model files for use in the empanada-dl python package and empanada-napari plugin. MitoNet_v1 is a 55 million parameter Panoptic-DeepLab architecture and MitoNet_v1_mini is a 23 million parameter Panoptic=BiFPN architecture. Quantized
Ryan Conrad, Kedar Narayan
core   +1 more source

Application of YOLOv8 Architecture Optimized based on BiFPN in Leather Defect Recognition

Pige Kexue Yu Gongcheng
Traditional image processing methods are difficult to effectively deal with complex backgrounds and defects with different scales. This paper proposed a YOLOv8 architecture optimization strategy that integrates Bidirectional Feature Pyramid Network ...
Hao TANG, Faming CHEN, Zhipeng FENG, Lingzhi HE   +3 more
doaj   +1 more source

AI‐Driven Deep Learning Framework for Detecting Subtle Surface Defects on Wind Turbine Blades

Wind Energy, Volume 29, Issue 3, March 2026.
ABSTRACT Wind turbine blade surface defect detection is of great significance in ensuring the safety and operational efficiency of wind power systems. However, accurately detecting subtle and small‐scale defects remains challenging under complex imaging conditions.
Shoutu Li, Jiawei Pan, Ling Liu, Jin Zhao, Nianxi Yue, Zhen Wu   +5 more
wiley   +1 more source

OralSegNet: An Approach to Early Detection of Oral Disease Using Transfer Learning

Oral Diseases, Volume 32, Issue 3, Page 791-808, March 2026.
ABSTRACT Objective Deep learning‐based segmentation system is proposed that exploits three variants of YOLOv11 architecture, namely YOLOv11n‐seg, YOLOv11s‐seg, and YOLOv11m‐seg for automated detection and localization of the oral disease conditions from photographic intraoral images.
Pranta Barua, Md Rakibul Islam, Mahmud Uz Zaman, Adel Alenazi, Fawaz Alqahtani, Huda Abutayyem, Maher Al Shayeb, Ruba Odeh, Nesrine A. Elsahn, Mohammad Khursheed Alam   +9 more
wiley   +1 more source

Multi‐Head Bias Fusion and Adaptive Aligned Gradient of Multi‐Task Learning for Joint Human Detection, Segmentation and Pose Estimation

IET Image Processing, Volume 20, Issue 1, January/December 2026.
YOLODSP incorporates multi‐task heads and introduces a BiasFusion module to predict the offsets of pose estimation and segmentation. With an overall average precision difference of no more than 1% relative to the combination of multiple single‐task methods, YOLODSP reduces the computational load by 35.2%, 38.8% and 40.5% on the YOLOv8‐nano, YOLOv8 ...
Feng Lu, Peng Ye, Yaowen Zhang, Ning Lu, Jinsheng Chen, Youchun Xu   +5 more
wiley   +1 more source

AHD‐YOLO: An Adaptive Hybrid Dynamic Network for Building Damage Detection

IET Image Processing, Volume 20, Issue 1, January/December 2026.
To address the issues of limited detection accuracy and high computational resource consumption in current deep learning‐based building damage detection, we propose a novel framework, AHD YOLO, built upon YOLOv11. AHD YOLO achieves an optimal balance between detection performance and computational resource efficiency, demonstrating strong potential for
Min Li, Tao Xu, Yinping Jiang, Peiyong Ji, Jingqi Hu, Wenlong Lui, Xuejian Ji, Ruiqiang Guo   +7 more
wiley   +1 more source