相关参考文献
注意：仅列出部分参考文献，下载原文获取全部文献信息。
Article
Environmental Sciences
Gensheng Hu et al.
Summary: This study presents a method for detecting diseased pine trees using unmanned aerial vehicle (UAV) and deep learning technology. The proposed method, which combines MobileNet for background removal and Faster R-CNN for distinguishing between diseased and healthy trees, outperforms traditional machine learning approaches and other deep learning methods in experiments.
GEOCARTO INTERNATIONAL
(2022)
Review
Computer Science, Artificial Intelligence
Yujian Mo et al.
Summary: This paper reviews the state-of-the-art technologies of semantic segmentation based on deep learning and investigates related works on weakly-supervised, domain adaptation, multi-modal data fusion, and real-time semantic segmentation.
Article
Environmental Sciences
Jie You et al.
Summary: This paper addresses the detection of pine wilt disease, creating a new dataset using orthophotographs and proposing a two-stage detection method. They collected disease and negative samples using a convolutional neural network and localized the disease using an object detection model. The system performance was improved using image augmentation methods, and achieved high accuracy in real-world tests.
Article
Environmental Sciences
Jixia Huang et al.
Summary: This paper proposes the use of deep convolutional neural networks (D-CNNs) for the detection of pine wood nematode disease in remote sensing images. Five popular models are trained using transfer learning on a dataset constructed from Gaofen-1 (GF-1) and Gaofen-2 (GF-2) images. The study shows that SqueezeNet performs better than other models in terms of transfer learning on the sample dataset. The model's recognition efficiency and accuracy are improved by referring to the Slim module structure, resulting in an improved model that can accurately and efficiently monitor pine wood nematode disease using remote sensing.
Article
Forestry
Zhao Sun et al.
Summary: This study used an unmanned aerial vehicle platform and multi-scale segmentation algorithm combined with object-oriented classification to successfully identify pine trees affected by pine wilt disease, providing a technical method for monitoring disease occurrence and identifying affected trees.
JOURNAL OF FORESTRY RESEARCH
(2022)
Article
Remote Sensing
Gensheng Hu et al.
Summary: This paper proposes a method for unmanned aerial vehicle remote sensing monitoring of pine forest diseases based on the improved Mask R-CNN. By adding improved modules, multi-scale receptive field block modules, and a multi-level fusion residual pyramid structure, the method enhances the detailed feature extraction of pine forest diseases and reduces background interference, thereby improving the accuracy of monitoring.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2022)
Article
Environmental Sciences
Jiahao Wang et al.
Summary: This study combines satellite remote sensing and deep learning technologies to investigate the accurate monitoring method for pine wilt disease (PWD). By constructing a dataset and selecting appropriate models, the study found that HRNet model performs the best in identifying discolored standing trees. By using a GAN generator for semi-supervised learning, the recognition accuracy can be further improved. In practical application, the method achieved a recall rate of 80.09% for PWD monitoring, which is of great significance for wide-scale monitoring, prevention, and control of PWD.
Article
Chemistry, Analytical
Zhengzhi Sun et al.
Summary: Pine wilt nematode disease, a rapidly spreading forest disease, can be controlled through the use of drone remote sensing to monitor infected pine trees. This study proposes an improved algorithm combining YOLOv4 and MobileNetv2 to enhance the accuracy and efficiency of identifying abnormal discoloration caused by pine wilt nematode disease.
Article
Environmental Sciences
Zemin Han et al.
Summary: The paper proposes a new detection method called multi-scale spatial supervision convolutional network (MSSCN) to identify standing dead trees (SDTs) in complex scenes based on airborne remote sensing imagery. The method utilizes a Gaussian kernel approach to generate a confidence map and a multi-scale spatial attention block to reduce the loss of spatial information. Experimental results show that the MSSCN model achieves high precision and recall rates, especially in pure pine forest type and low disease outbreak intensity areas.
Article
Chemistry, Multidisciplinary
Sulan Zhang et al.
Summary: In this study, a feasible and effective method using unmanned aerial vehicles and hyperspectral imaging technology was proposed to monitor and estimate the infection stage of pine wilt disease. By extracting spectral features and implementing genetic algorithm and support vector machine classifier, it achieved high accuracy in distinguishing different infection stages of pine trees.
APPLIED SCIENCES-BASEL
(2022)
Article
Environmental Sciences
Ayako H. Oide et al.
Summary: This study presents an optimal method for detecting pine wilt disease using a simple unmanned aerial vehicle (UAV)-mounted visible color sensor. The experimental results show that the combination of the HSV color space dataset and the artificial neural network algorithm achieves the highest detection accuracy (0.995), while logistic regression, linear support vector machine, support vector machine, and artificial neural network also have high detection accuracy in the RGB color space (0.990, 0.990, 0.990, 0.985 respectively) with low overfitting possibility.
REMOTE SENSING APPLICATIONS-SOCIETY AND ENVIRONMENT
(2022)
Article
Ecology
Wujian Ye et al.
Summary: This study addresses the issue of pine wilt disease detection by designing an automatic pest detection framework that combines aerial photography data from unmanned aerial vehicles and satellite remote sensing images. The performance of the developed model is optimized using data augmentation techniques. Experimental results show that the proposed model achieves a high recall rate in detecting pest-infested forest areas.
ECOLOGICAL INFORMATICS
(2022)
Article
Ecology
Gensheng Hu et al.
Summary: Accurate detection and classification of pine tree diseases are crucial for monitoring tree growth and controlling forest diseases. The proposed method combines DDYOLOv5 and ResNet50 networks to improve detection accuracy using preprocessing and specific modules. Experimental results demonstrate higher precision and recall compared to other methods.
ECOLOGICAL INFORMATICS
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Zhuang Liu et al.
Summary: The development of visual recognition has gone through stages from ConvNets to ViTs and then to hybrid approaches. In this work, the design of a pure ConvNet is reexamined and several key components are discovered, resulting in the construction of the ConvNeXt model series. These models compete with Transformers in terms of accuracy and performance while maintaining the simplicity and efficiency of ConvNets.
2022 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2022)
Article
Remote Sensing
Jingtao Li et al.
Summary: A novel PWD detection framework based on deep one-class classification (OCC) is proposed in this study, which utilizes three-dimensional (3D) convolutional layers and transformer blocks to extract features, achieving fine pixel-level PWD detection results.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2022)
Article
Geography, Physical
Zhuo Zang et al.
Summary: This study selected target and reference trees from a Chinese fir plantation and successfully detected the early dying process of damaged trees using hyperspectral data and the proposed GRRSGI index.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Article
Forestry
Run Yu et al.
Summary: This study systematically divided the Pine Wilt Disease (PWD) infection into four stages and used a combination of UAV-based multispectral images and target detection algorithms to identify infected pine trees, providing technical support for early monitoring of PWD and obtaining the distribution of infected trees.
FOREST ECOLOGY AND MANAGEMENT
(2021)
Article
Environmental Sciences
Jun Qin et al.
Summary: This paper introduces a new network, SCANet, based on UAV multi-spectral remote sensing images to identify pine nematode disease. The proposed method achieved an overall accuracy rate of 79% with high precision and recall values, outperforming other existing methods. It provides a fast, precise, and practical approach for identifying nematode disease and supports the surveillance and control of this destructive disease effectively.
Article
Environmental Sciences
Run Yu et al.
Summary: The study demonstrates that using the 3D-Res CNN model can more accurately and effectively detect early infected pine trees in hyperspectral images. This method outperforms current early detection methods based on hyperspectral images for PWD and can be applied to detect pine trees damaged by other diseases or insect pests in the forest.
Article
Forestry
Bizhi Wu et al.
Summary: This study utilizes unmanned aerial vehicles to collect images for early diagnosis of pine wilt disease. Through the application of deep learning algorithms, particularly the YOLOv3 model, the research demonstrates improved processing speed and accuracy for early detection of the disease.
FOREST ECOLOGY AND MANAGEMENT
(2021)
Article
Plant Sciences
Xinggang Tang et al.
Summary: Pine wilt disease caused by the pinewood nematode is a devastating forest disease in China, with the southern area of the Yangtze River being the most severely affected. Climate change influences the distribution of pine trees and the spread of the nematode. In the future, the suitable habitat for the nematode is expected to increase, requiring appropriate measures to prevent its expansion.
FRONTIERS IN PLANT SCIENCE
(2021)
Article
Remote Sensing
Run Yu et al.
Summary: The study divided pine wilt disease (PWD) infection into five stages and integrated hyperspectral imaging with LiDAR data for detection. The combined use of hyperspectral and LiDAR data showed the best accuracies in predicting PWD infection stages, improving monitoring and control. LiDAR metrics were essential in the classification model, highlighting the importance of combining these two data sources for precise prediction at the tree level.
INTERNATIONAL JOURNAL OF APPLIED EARTH OBSERVATION AND GEOINFORMATION
(2021)
Article
Environmental Sciences
Guang Li et al.
Summary: This study proposed a multi-spectral image processing method for extracting information on sunflower lodging by improving deep learning methods. The results showed that deep learning methods outperformed traditional random forest methods in terms of accuracy and information extraction, and adding NIR bands helped improve the accuracy of lodging information.
Article
Environmental Sciences
Biyao Zhang et al.
Summary: This study proposes a spatiotemporal change detection method to reduce false detections in tree-scale pine wilt disease monitoring in complex landscapes. The method captures spectral, temporal, and spatial features through calculating the Normalized Green-Red Difference Index, contrasting NGRDI images of similar dates in adjacent years, and enhancing the change index using a convolution kernel. The proposed method effectively distinguishes wilted pine trees from other easily confused objects, improving user's accuracy while maintaining the same level of producer's accuracy compared to single-date image classification.
Review
Environmental Sciences
Yago Diez et al.
Summary: This article highlights the importance of forests to the planet and their current applications in forestry research, focusing on studies that use deep learning and RGB images collected by UAVs to address practical forestry research problems. The article summarizes the strengths and methodological issues of existing studies, and provides public data and code resources to aid researchers interested in working in this area.
Article
Environmental Sciences
Lang Xia et al.
Summary: The study found that using deep learning techniques combined with UAV images can effectively identify and evaluate the degree of infection in pine trees. DeepLabv3+ demonstrated the best performance, showing higher accuracy and finer boundaries.
Article
Remote Sensing
Huan Tao et al.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2020)
Article
Engineering, Multidisciplinary
Mutiara Syifa et al.
Article
Environmental Sciences
Marian-Daniel Iordache et al.
Article
Plant Sciences
Kenichi Yazaki et al.
JOURNAL OF EXPERIMENTAL BOTANY
(2018)
Article
Forestry
A. J. Carnegie et al.
AUSTRALIAN FORESTRY
(2018)
Article
Forestry
Makihiko Ikegami et al.
FOREST ECOLOGY AND MANAGEMENT
(2018)
Article
Forestry
So-Ra Kim et al.
Review
Microbiology
Diogo N. Proenca et al.
Article
Computer Science, Artificial Intelligence
Evan Shelhamer et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2017)
Article
Computer Science, Artificial Intelligence
Shaoqing Ren et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2017)
Article
Multidisciplinary Sciences
Akiko Hirata et al.
Proceedings Paper
Computer Science, Artificial Intelligence
Kaiming He et al.
2017 IEEE INTERNATIONAL CONFERENCE ON COMPUTER VISION (ICCV)
(2017)
Review
Multidisciplinary Sciences
Yann LeCun et al.
Article
Engineering, Electrical & Electronic
Asim Anees et al.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2014)
Article
Geography, Physical
Elfatih M. Abdel-Rahman et al.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2014)
Article
Zoology
David Hunt
Article
Remote Sensing
J. C. White et al.
INTERNATIONAL JOURNAL OF REMOTE SENSING
(2007)
Article
Forestry
MA Wulder et al.
FOREST ECOLOGY AND MANAGEMENT
(2006)
