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
Agriculture, Multidisciplinary
Angelo Cardellicchio et al.
Summary: Plant phenotyping is the study of complex plant traits, and human operators often carry out evaluations that may be biased by their experience and skill. With the development of computer vision systems, artificial intelligence algorithms play a crucial role in the automation and quantitative analysis of large amounts of data.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2023)
Article
Multidisciplinary Sciences
Xinjie Wei et al.
Summary: This study proposes a Shine-Muscat Grape Detection Model (S-MGDM) based on improved YOLOv3 to enhance the detection accuracy of green grapes. The model incorporates DenseNet, depth-separable convolution, CBAM, SPPNet, PANet, and FPN to extract grape features, increase perceptual field, and improve inter-network information flow. With the use of CIOU regression loss function and modified frame size, the improved model achieves higher accuracy and faster detection speed compared to the original network model. It shows excellent detection accuracy and real-time performance, providing valuable reference for accurate identification of ripe Shine-Muscat grapes.
SCIENTIFIC REPORTS
(2023)
Article
Green & Sustainable Science & Technology
Emmanouil Tziolas et al.
Summary: The increased cost of labor in modern viticulture, coupled with labor shortages, poses challenges. Collaborative robots can potentially solve these issues and reduce operational expenses in agriculture. This study evaluates the economic viability of using collaborative robots compared to conventional labor in grape cultivation in Greece. The results show the significant potential of robots in specific viticultural operations, such as weed control and pruning. However, certain operations like defoliation and tying were less efficient with robots. Overall, the adoption of collaborative robots can reduce costs and address labor shortages, contributing to sustainable agriculture.
Article
Engineering, Marine
Yuqing Liu et al.
Summary: This paper proposes a lightweight object detection network Tuna-YOLO for mobile devices, which achieves accurate detection of tuna through improved network structure and knowledge distillation. Experimental results show that Tuna-YOLO outperforms other versions of YOLO network in terms of parameter quantity, detection accuracy, and calculation speed. This research is of great significance for subsequent deployment of algorithms on mobile devices and real-time catch statistics.
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2023)
Article
Plant Sciences
Weishi Xu et al.
Summary: This paper proposes an accurate and lightweight model for apple leaf disease detection based on YOLO-V5s (ALAD-YOLO). The model achieves a good balance between detection speed and accuracy in apple leaf disease detection. By collecting an apple leaf disease detection dataset and applying various data augmentation algorithms, the generalization of the model is improved. Experimental results show that ALAD-YOLO achieves a detection accuracy of 90.2% on the test set and reduces the floating point of operations (FLOPs) to 6.1 G. Overall, this paper provides an accurate and efficient detection method for apple leaf disease detection and other related fields.
FRONTIERS IN PLANT SCIENCE
(2023)
Article
Agronomy
Congyue Wang et al.
Summary: In order to improve the efficiency of mechanical automatic picking of cherry tomatoes in a precision agriculture environment, this study proposes an improved target detection algorithm based on YOLOv5n. By updating the anchor box, introducing the coordinate attention mechanism, and replacing the loss function, the performance of the detection algorithm is enhanced. Experimental results show that the improved model achieves a 1.4% increase in both precision and recall, with an average accuracy of 95.2% and a detection time of 5.3 ms. It is suitable for real-time detection in embedded systems and mobile devices, providing rapid and accurate target recognition guidance for mechanical automatic picking.
Proceedings Paper
Computer Science, Artificial Intelligence
Chien-Yao Wang et al.
Summary: Real-time object detection is an important research topic in computer vision, and the development of new approaches in architecture optimization and training optimization has led to two related research topics. To address these topics, a trainable solution combining flexible and efficient training tools, proposed architecture, and compound scaling method is proposed. YOLOv7 outperforms all known object detectors in terms of speed and accuracy, achieving the highest AP accuracy of 56.8% among real-time object detectors with 30 FPS or higher on GPU V100.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR
(2023)
Proceedings Paper
Computer Science, Theory & Methods
Vassilis G. Kaburlasos et al.
Summary: Accurately predicting agricultural yield is crucial for resource allocation. This study proposes a method based on fuzzy set theory to interpret the estimated cumulative distribution function as a cumulative possibility distribution for predicting the distribution of agricultural yield. Experimental results demonstrate the feasibility of this method.
APPLICATIONS OF FUZZY TECHNIQUES, NAFIPS 2022
(2023)
Article
Environmental Sciences
Vasilis Psiroukis et al.
Summary: The aim of this study was to automate the maturity classification of broccoli using object detection architectures and UAV images, and evaluate their effectiveness. The results showed that the automated approach performed well, especially with the best-performing architectures.
Article
Agronomy
Marco Sozzi et al.
Summary: This paper evaluates six versions of the YOLO object detection algorithm for real-time bunch detection and counting in grapes. The results show that YOLOv4-tiny achieves the best combination of accuracy and speed, making it suitable for real-time grape yield estimation.
Article
Agronomy
Xinguang Wei et al.
Summary: A plant phenotype-monitoring platform and back-propagation neural network algorithm were used to predict grape maturity in solar greenhouses. By measuring and evaluating the physical and chemical properties of grapes and the color values of the grape skin, the maturity of grapes was successfully predicted and divided into two stages. The combination of predictive factors showed variation in accuracy for different grape varieties.
Article
Agriculture, Multidisciplinary
Xuewen Wang et al.
Summary: A lightweight small object detection architecture called LDS-YOLO is proposed in this paper, with a novel feature extraction module, SPP with SoftPool method, and depth-wise separable convolution introduced to address the challenge of identifying dead trees efficiently. The experimental results show that the proposed method performs well compared to state-of-the-art models, with a high AP of 89.11% and a small parameter size of 7.6 MB.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2022)
Article
Chemistry, Multidisciplinary
Mei Jia et al.
Summary: This study investigates the quality classification of the Red Globe grape using computer vision and deep learning. A framework called feature normalization reweighting regression (FNRR) is proposed to address the imbalanced distribution of sugar content of the grape datasets. The experimental results show that the FNRR framework can accurately measure the sugar content of grapes using convolution neural networks and a visual transformer model.
APPLIED SCIENCES-BASEL
(2022)
Article
Computer Science, Information Systems
Chang Qiu et al.
Summary: A grape maturity detection and visual pre-positioning algorithm based on improved YOLOv4 was proposed in this study to guide grape picking robots in recognizing and classifying grapes with different maturity levels in complex orchard environments. The algorithm achieved high accuracy and detection speed, with an overall average accuracy of 93.52% and an average detection time of 10.82 ms. The pre-positioning method based on binocular stereo vision obtained spatial position information of grape clusters with low error rates, making it reliable for precise grape picking in orchard environments.
Proceedings Paper
Automation & Control Systems
Ruth A. Bastidas-Alva et al.
Summary: This research aims to recognize and classify the ripening state of Trinitario cocoa using the YOLO-v5 artificial vision technique. The method involves preprocessing, processing, and post-processing, including data acquisition, annotation, and augmentation. The trained model achieved high accuracy in real-time recognition and classification of Trinitario cocoa based on its ripening stage.
2022 ASIA CONFERENCE ON ADVANCED ROBOTICS, AUTOMATION, AND CONTROL ENGINEERING (ARACE 2022)
(2022)
Article
Agricultural Engineering
Dandan Wang et al.
Summary: This study developed an accurate method for detecting apple fruitlets based on a channel pruned YOLO V5s deep learning algorithm, with a small model size. Experimental results showed that the method effectively detected apple fruitlets under different conditions, outperforming seven other methods.
BIOSYSTEMS ENGINEERING
(2021)
Article
Computer Science, Information Systems
Rafael Padilla et al.
Summary: This study provides an overview of evaluation methods used in object detection competitions, examines the influence of different annotation formats on evaluation results, and offers an open-source toolkit supporting various annotation formats and performance metrics for researchers to evaluate their detection algorithms. Furthermore, it introduces a new metric for evaluating object detection in videos based on spatio-temporal overlap.
Review
Geography, Physical
Teja Kattenborn et al.
Summary: Identifying and characterizing vascular plants in time and space is essential for various disciplines such as forestry, conservation, and agriculture. Remote sensing technology, specifically utilizing deep learning methods like Convolutional Neural Networks (CNN), is crucial in improving efficiency and accuracy in extracting vegetation properties from remote sensing imagery. The modularity and flexibility of CNN frameworks allow for adaptation to different architectures, particularly for multi-modal or multi-temporal applications, ultimately leading to a new era of vegetation remote sensing.
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING
(2021)
Article
Chemistry, Analytical
Cinmayii A. Garillos-Manliguez et al.
Summary: This study proposes a novel multimodal classification method for estimating fruit maturity, utilizing deep convolutional neural networks and data acquired from visible-light and hyperspectral imaging systems. By modifying existing architectures and analyzing sensitivity through multimodal data cubes, the multimodal variants can achieve high accuracy in classifying six stages of fruit maturity. This indicates the great potential of multimodal deep learning architectures and imaging for real-time in-field fruit maturity estimation and industrial applications.
Article
Computer Science, Information Systems
Eleni Vrochidou et al.
Summary: This study aims to extend Industry 4.0 practices to agriculture, with a focus on fruit harvesting, particularly grapes. The Autonomous Robot for Grape harvesting (ARG) features an integrated system architecture consisting of aerial, remote-control, and ground units, designed for various viticultural operations such as harvest, green harvest, and defoliation. The modular design allows for potential extension to different crops and orchards.
Article
Environmental Studies
David Christian Rose et al.
Summary: This article emphasizes the importance of incorporating social sustainability into agricultural revolutions, proposing a framework of multi-actor co-innovation to guide responsible socio-technical transitions. By increasing the involvement of people in agricultural innovation systems and following responsible innovation principles, the likelihood of achieving social sustainability in this technological revolution can be enhanced.
Article
Instruments & Instrumentation
Syed Sohaib Ali Shah et al.
INFRARED PHYSICS & TECHNOLOGY
(2020)
Review
Imaging Science & Photographic Technology
Efthimia Mavridou et al.
JOURNAL OF IMAGING
(2019)
Review
Agriculture, Multidisciplinary
Andreas Kamilaris et al.
COMPUTERS AND ELECTRONICS IN AGRICULTURE
(2018)
Review
Food Science & Technology
Amelie Rabot et al.
