Water Column Detection Method at Impact Point Based on Improved YOLOv4 Algorithm DOI Open Access
Jiaowei Shi, Shiyan Sun,

Zhangsong Shi

et al.

Sustainability, Journal Year: 2022, Volume and Issue: 14(22), P. 15329 - 15329

Published: Nov. 18, 2022

For a long time, the water column at impact point of naval gun firing sea has mainly depended on manual detection methods for locating, which problems such as low accuracy, subjectivity and inefficiency. In order to solve above problems, this paper proposes method based an improved you-only-look-once version 4 (YOLOv4) algorithm. Firstly, detects antenna through Hoffman line constrain sensitive area in current image so improve accuracy detection; secondly, density-based spatial clustering applications with noise (DBSCAN) + K-means algorithm is used obtain better prior bounding box, input into YOLOv4 network positioning column; finally, convolutional block attention module (CBAM) added PANet structure column. The experimental results show that can effectively point.

Language: Английский

A deep learning-based adaptive denoising approach for fine identification of rock microcracks from noisy strain data DOI
Shuai Zhao,

Dian-Rui Mu,

Dao-Yuan Tan

et al.

Engineering Applications of Artificial Intelligence, Journal Year: 2025, Volume and Issue: 148, P. 110471 - 110471

Published: March 11, 2025

Language: Английский

Citations

1
Deep-reinforcement-learning-based UAV autonomous navigation and collision avoidance in unknown environments DOI Creative Commons
Fei Wang, Xiaoping Zhu, Zhou Zhou

et al.

Chinese Journal of Aeronautics, Journal Year: 2023, Volume and Issue: 37(3), P. 237 - 257

Published: Oct. 16, 2023

In some military application scenarios, Unmanned Aerial Vehicles (UAVs) need to perform missions with the assistance of on-board cameras when radar is not available and communication interrupted, which brings challenges for UAV autonomous navigation collision avoidance. this paper, an improved deep-reinforcement-learning algorithm, Deep Q-Network a Faster R-CNN model Data Deposit Mechanism (FRDDM-DQN), proposed. A (FR) introduced optimized obtain ability extract obstacle information from images, new replay memory (DDM) designed train agent better performance. During training, two-part training approach used reduce time spent on as well retraining scenario changes. order verify performance proposed method, series experiments, including test typical episodes conducted in 3D simulation environment. Experimental results show that trained by FRDDM-DQN has navigate autonomously avoid collisions, performs compared FR-DQN, FR-DDQN, FR-Dueling DQN, YOLO-based YDDM-DQN, original FR output-based FR-ODQN.

Language: Английский

Citations

20
Coal-rock interface real-time recognition based on the improved YOLO detection and bilateral segmentation network DOI Creative Commons

Shuzhan Xu,

Wanming Jiang,

Quansheng Liu

et al.

Underground Space, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 1, 2024

Language: Английский

Citations

5
Automatic borehole fracture detection and characterization with tailored Faster R-CNN and simplified Hough transform DOI

Shuyang Han,

Xiao Xiao,

Benyang Song

et al.

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 126, P. 107024 - 107024

Published: Sept. 2, 2023

Language: Английский

Citations

12
A dynamic weighted feature fusion lightweight algorithm for safety helmet detection based on YOLOv8 DOI

Hongge Ren,

Anni Fan, Jian Zhao

et al.

Measurement, Journal Year: 2025, Volume and Issue: unknown, P. 117572 - 117572

Published: April 1, 2025

Language: Английский

Citations

0
Deep learning in alloy material microstructures: Application and prospects DOI
Lun Che, Zhongping He, Kaiyuan Zheng

et al.

Materials Today Communications, Journal Year: 2023, Volume and Issue: 37, P. 107531 - 107531

Published: Nov. 9, 2023

Language: Английский

Citations

10
Insulation aging condition assessment of transformer in the visual domain based on SE-CNN DOI
Aniket Vatsa, Ananda Shankar Hati

Engineering Applications of Artificial Intelligence, Journal Year: 2023, Volume and Issue: 128, P. 107409 - 107409

Published: Nov. 14, 2023

Language: Английский

Citations

7
Design of fine motion control system for aerobics athletes based on light imaging equipment detection and image processing technology DOI
Liu Ai

Optical and Quantum Electronics, Journal Year: 2024, Volume and Issue: 56(4)

Published: Jan. 31, 2024

Language: Английский

Citations

2
Improved Lightweight YOLOv4 Foreign Object Detection Method for Conveyor Belts Combined with CBAM DOI Creative Commons

Jiehui Liu,

Hongchao Qiao, Lijie Yang

et al.

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(14), P. 8465 - 8465

Published: July 21, 2023

During the operation of belt conveyor, foreign objects such as large gangue and anchor rods may be mixed into conveyor belt, resulting in tears fractures, which affect transportation efficiency production safety. In this paper, we propose a lightweight target detection algorithm, GhostNet-CBAM-YOLOv4, to resolve problem difficulty detecting at high-speed movement an underground belt. The Kmeans++ clustering method was used preprocess data set obtain box suitable for object size. GhostNet module replaced backbone network, reducing model’s parameters. CBAM attention introduced enhance ability feature extraction facing complex environment under mine. depth separable convolution simplify model structure reduce number parameters calculations. accuracy improved on body reached 99.32%, rate 54.7 FPS, 6.83% 42.1% higher than original YOLOv4 model, respectively. performed better other two datasets could effectively avoid misdetection omission detection. comparison experiments with similar methods, our proposed also demonstrated good performance, verifying its effectiveness.

Language: Английский

Citations

6
Body Weight Estimation for Pigs Based on 3D Hybrid Filter and Convolutional Neural Network DOI Creative Commons
Zihao Liu, Jingyi Hua, Hongxiang Xue

et al.

Sensors, Journal Year: 2023, Volume and Issue: 23(18), P. 7730 - 7730

Published: Sept. 7, 2023

The measurement of pig weight holds significant importance for producers as it plays a crucial role in managing growth, health, and marketing, thereby facilitating informed decisions regarding scientific feeding practices. On one hand, the conventional manual weighing approach is characterized by inefficiency time consumption. other has potential to induce heightened stress levels pigs. This research introduces hybrid 3D point cloud denoising precise estimation. By integrating statistical filtering DBSCAN clustering techniques, we mitigate estimation bias overcome limitations feature extraction. convex hull technique refines dataset pig's back, while voxel down-sampling enhances real-time efficiency. Our model integrates back parameters with convolutional neural network (CNN) accurate Experimental analysis indicates that mean absolute error (MAE), percent (MAPE), root square (RMSE) proposed this are 12.45 kg, 5.36%, 12.91 respectively. In contrast currently available methods based on 2D suggested offers advantages simplified equipment configuration reduced data processing complexity. These benefits achieved without compromising accuracy Consequently, method presents an effective monitoring solution management, leading human resource losses improved welfare breeding.

Language: Английский

Citations

6