Exploiting 2D Neural Network Frameworks for 3D Segmentation Through Depth Map Analytics of Harvested Wild Blueberries (Vaccinium angustifolium Ait.) DOI Creative Commons

Connor C. Mullins,

Travis J. Esau, Qamar uz Zaman

et al.

Journal of Imaging, Journal Year: 2024, Volume and Issue: 10(12), P. 324 - 324

Published: Dec. 15, 2024

This study introduced a novel approach to 3D image segmentation utilizing neural network framework applied 2D depth map imagery, with Z axis values visualized through color gradation. research involved comprehensive data collection from mechanically harvested wild blueberries populate and red–green–blue (RGB) images of filled totes time-of-flight RGB cameras, respectively. Advanced models the YOLOv8 Detectron2 frameworks were assessed for their capabilities. Notably, models, particularly YOLOv8n-seg, demonstrated superior processing efficiency, an average time 18.10 ms, significantly faster than which exceeded 57 while maintaining high performance mean intersection over union (IoU) 0.944 Matthew’s correlation coefficient (MCC) 0.957. A qualitative comparison masks indicated that YOLO produced smoother more accurate object boundaries, whereas showed jagged edges under-segmentation. Statistical analyses, including ANOVA Tukey’s HSD test (α = 0.05), confirmed on maps (p < 0.001). concludes by recommending YOLOv8n-seg model real-time in precision agriculture, providing insights can enhance volume estimation, yield prediction, resource management practices.

Language: Английский

YOLOv8: A Novel Object Detection Algorithm with Enhanced Performance and Robustness DOI

Rejin Varghese,

M. Sambath

Published: April 18, 2024

In recent years, the You Only Look Once (YOLO) series of object detection algorithms have garnered significant attention for their speed and accuracy in real-time applications. This paper presents YOLOv8, a novel algorithm that builds upon advancements previous iterations, aiming to further enhance performance robustness. Inspired by evolution YOLO architectures from YOLOv1 YOLOv7, as well insights comparative analyses models like YOLOv5 YOLOv6, YOLOv8 incorporates key innovations achieve optimal accuracy. Leveraging mechanisms dynamic convolution, introduces improvements specifically tailored small detection, addressing challenges highlighted YOLOv7. Additionally, integration voice recognition techniques enhances algorithm's capabilities video-based demonstrated The proposed undergoes rigorous evaluation against state-of-the-art benchmarks, showcasing superior terms both computational efficiency. Experimental results on various datasets confirm effectiveness across diverse scenarios, validating its suitability real-world contributes ongoing research presenting versatile high-performing algorithm, poised address evolving needs computer vision systems.

Language: Английский

Citations

213
Applying optimized YOLOv8 for heritage conservation: enhanced object detection in Jiangnan traditional private gardens DOI Creative Commons
Chan Gao, Qingzhu Zhang,

Zheyu Tan

et al.

Heritage Science, Journal Year: 2024, Volume and Issue: 12(1)

Published: Jan. 29, 2024

Abstract This study aims to promote the protection and inheritance of cultural heritage in private gardens Jiangnan area China. By establishing a precise visual labeling system accelerating construction database for garden features, we deepen understanding design philosophy. To this end, propose an improved recognition model based on You Only Look Once (YOLO) v8. is particularly suitable processing environments with characteristics such as single or complex structures, rich depth field, cluttered targets, effectively enhancing accuracy efficiency object recognition. integrates Diverse Branch Block (DBB), Bidirectional Feature Pyramid Network (BiFPN), Dynamic Head modules (DyHead) optimize accuracy, feature fusion, detection representational capability, respectively. The enhancements elevated model's by 8.7%, achieving mean average precision ([email protected]) value 57.1%. A specialized dataset, comprising 4890 images encapsulating various angles lighting conditions gardens, was constructed realize this. Following manual annotation application diverse data augmentation strategies, dataset bolsters generalization robustness model. Experimental outcomes reveal that, compared its predecessor, has witnessed increments 15.16%, 3.25%, 11.88% precision, mAP0.5, mAP0.5:0.95 metrics, respectively, demonstrating exemplary performance real-time target elements. research not only furnishes robust technical support digitization intelligent but also provides potent methodological reference classification analogous domains.

Language: Английский

Citations

13
Thermal Canopy Segmentation in Tomato Plants: A Novel Approach with Integration of YOLOv8-C and FastSAM DOI Creative Commons

P. Hemamalini,

Chandraprakash MK,

RH Laxman

et al.

Smart Agricultural Technology, Journal Year: 2025, Volume and Issue: unknown, P. 100806 - 100806

Published: Jan. 1, 2025

Language: Английский

Citations

1
Rapeseed Flower Counting Method Based on GhP2-YOLO and StrongSORT Algorithm DOI Creative Commons
Nan Wang,

Haijuan Cao,

Xia Huang

et al.

Plants, Journal Year: 2024, Volume and Issue: 13(17), P. 2388 - 2388

Published: Aug. 27, 2024

Accurately quantifying flora and their respective anatomical structures within natural ecosystems is paramount for both botanical breeders agricultural cultivators. For breeders, precise plant enumeration during the flowering phase instrumental in discriminating genotypes exhibiting heightened frequencies, while growers, such data inform potential crop rotation strategies. Moreover, quantification of specific components, as flowers, can offer prognostic insights into yield variances among different genotypes, thereby facilitating informed decisions pertaining to production levels. The overarching aim present investigation explore capabilities a neural network termed GhP2-YOLO, predicated on advanced deep learning techniques multi-target tracking algorithms, specifically tailored rapeseed flower buds blossoms from recorded video frames. Building upon foundation renowned object detection model YOLO v8, this integrates specialized P2 head Ghost module augment model's capacity detecting diminutive targets with lower resolutions. This modification not only renders more adept at target identification but also it lightweight less computationally intensive. optimal iteration GhP2-YOLOm demonstrated exceptional accuracy samples, showcasing an impressive mean average precision 50% intersection over union metric surpassing 95%. Leveraging virtues StrongSORT, subsequent blossom patterns dataset was adeptly realized. By selecting 20 segments comparative analysis between manual automated counts buds, overall count, robust correlation evidenced, R-squared coefficients measuring 0.9719, 0.986, 0.9753, respectively. Conclusively, user-friendly "Rapeseed detection" system developed utilizing GUI PyQt5 interface, visualization flowers buds. holds promising utility field surveillance apparatus, enabling agriculturalists monitor developmental progress real time. innovative study introduces tallying methodologies footage, positioning convolutional networks protocols invaluable assets realms research administration.

Language: Английский

Citations

7
High-resolution density assessment assisted by deep learning of Dendrophyllia cornigera (Lamarck, 1816) and Phakellia ventilabrum (Linnaeus, 1767) in rocky circalittoral shelf of Bay of Biscay DOI Creative Commons

Alberto Gayá-Vilar,

Adolfo Cobo, Alberto Abad‐Uribarren

et al.

PeerJ, Journal Year: 2024, Volume and Issue: 12, P. e17080 - e17080

Published: March 7, 2024

This study presents a novel approach to high-resolution density distribution mapping of two key species the 1170 “Reefs” habitat, Dendrophyllia cornigera and Phakellia ventilabrum , in Bay Biscay using deep learning models. The main objective this was establish pipeline based on models extract data from raw images obtained by remotely operated towed vehicle (ROTV). Different object detection were evaluated compared various shelf zones at head submarine canyon systems metrics such as precision, recall, F1 score. best-performing model, YOLOv8, selected for generating maps high spatial resolution. also generated synthetic augment training assess generalization capacity proposed provides cost-effective non-invasive method monitoring assessing status these important reef-building their habitats. results have implications management protection habitat Spain other marine ecosystems worldwide. These highlight potential improve efficiency accuracy vulnerable ecosystems, allowing informed decisions be made that can positive impact conservation.

Language: Английский

Citations

6
Advancing architectural heritage: precision decoding of East Asian timber structures from Tang dynasty to traditional Japan DOI Creative Commons
Chan Gao,

Genfeng Zhao,

Sen Gao

et al.

Heritage Science, Journal Year: 2024, Volume and Issue: 12(1)

Published: June 28, 2024

Abstract The convergence of cultural and aesthetic elements in timber structures from China’s Tang Dynasty (618–907 AD) traditional Japanese architecture provides a rich tapestry architectural evolution cross-cultural exchanges. Addressing the challenge distinguishing understanding intricate styles these is significant for both historical comprehension preservation efforts. This research introduces an innovative approach by integrating Multi-Head Attention (MHA) mechanism into YOLOv8 model, enhancing detection features with improved precision recall. Our novel YOLOv8-MHA model not only demonstrates notable improvement recognizing details but also significantly advances state art object within complex settings. Quantitative results underscore model’s effectiveness, achieving 95.6%, recall 85.6%, mean Average Precision (mAP@50) 94% across various Intersection over Union (IoU) thresholds. These metrics highlight superior capability to accurately identify classify elements, especially environments nuanced details, utilizing enhanced algorithm. application our extends beyond mere analysis; it offers new insights interplay identity adaptability inherent East Asian heritage. study establishes solid foundation meticulous classification analysis expansive context, thereby enriching traditions.

Language: Английский

Citations

4
Novel Fusion Technique for High-Performance Automated Crop edge Detection in Smart Agriculture DOI Creative Commons
F. Martínez, James Brian Romaine, Princy Johnson

et al.

IEEE Access, Journal Year: 2025, Volume and Issue: 13, P. 22429 - 22445

Published: Jan. 1, 2025

Language: Английский

Citations

0
Using YOLOv8 for Building Damage Identification in Japan's Noto Region Following Earthquakes: A Deep Learning-Based Approach DOI
Chan Gao,

Genfeng Zhao,

Sen Gao

et al.

Lecture notes in civil engineering, Journal Year: 2025, Volume and Issue: unknown, P. 37 - 48

Published: Jan. 1, 2025

Language: Английский

Citations

0
Recognition and phenotypic detection of maize stem and leaf at seedling stage based on 3D reconstruction technique DOI

Haiou Guan,

Xueyan Zhang, Xiaodan Ma

et al.

Optics & Laser Technology, Journal Year: 2025, Volume and Issue: 187, P. 112787 - 112787

Published: March 14, 2025

Language: Английский

Citations

0
Plant recognition and counting of Amorphophallus konjac based on UAV RGB imagery and deep learning DOI
Ziyi Yang,

Kunrong Hu,

Weili Kou

et al.

Computers and Electronics in Agriculture, Journal Year: 2025, Volume and Issue: 235, P. 110352 - 110352

Published: March 30, 2025

Language: Английский

Citations

0