Computers and Electronics in Agriculture, Journal Year: 2024, Volume and Issue: 227, P. 109498 - 109498
Published: Nov. 1, 2024
Language: Английский
Computers and Electronics in Agriculture, Journal Year: 2025, Volume and Issue: 233, P. 110152 - 110152
Published: March 4, 2025
Language: Английский
Citations
0
Food Innovation and Advances, Journal Year: 2025, Volume and Issue: 4(1), P. 108 - 115
Published: Jan. 1, 2025
Language: Английский
Citations
0
Artificial Intelligence in Agriculture, Journal Year: 2024, Volume and Issue: 12, P. 97 - 108
Published: May 26, 2024
Mapping hazelnut orchards can facilitate land planning and utilization policies, supporting the development of cooperative precision farming systems. The present work faces detection crops using optical radar remote sensing data. A comparative study Machine Learning techniques is presented. system proposed utilizes multi-temporal data from Sentinel-1 Sentinel-2 datasets extracted over several years processed with cloud tools. We provide a dataset 62,982 labeled samples, 16,561 samples belonging to 'hazelnut' class 46,421 'other' class, collected in 8 heterogeneous geographical areas Viterbo province. Two different tests are conducted: firstly, we use Nested 5-Fold Cross-Validation methodology train, optimize, compare algorithms on single area. In second experiment, were trained one area tested remaining seven areas. developed demonstrates how AI analysis applied valid technology for mapping. From results, it emerges that Random Forest classifier highest generalizability, achieving best performance test an accuracy 96% F1 score 91% class.
Language: Английский
Citations
3
Remote Sensing, Journal Year: 2024, Volume and Issue: 16(20), P. 3780 - 3780
Published: Oct. 11, 2024
Employing drones and hyperspectral imagers for large-scale, precise evaluation of nitrogen (N) concentration in Carya cathayensis Sarg canopies is crucial accurately managing fertilization C. cultivation. This study gathered five sets imagery data from plantations across four distinct locations with varying environmental stresses using drones. The research assessed the canopy trees both during singular growth periods throughout their entire cycles. objective was to explore influence band combinations spectral index formula configurations on predictive capability indices (HIs) N (CNC), optimize performance between HIs machine learning approaches, validate efficacy optimized HI algorithms. findings revealed following: (i) Optimized demonstrated optimal full cycles Sarg. most effective model optimized–modified–normalized difference vegetation (opt-mNDVI), achieving an adjusted coefficient determination (R2) 0.96 a root mean square error (RMSE) 0.71. For cycle, model, also opt-mNDVI, attained R2 0.75 RMSE 2.11; (ii) substantially enhanced HIs’ by 16% 71%, while choice three-band two-band influenced capacity 4% 46%. Hence, utilizing combined Unmanned Aerial Vehicle (UAV) imaging evaluate under complex field conditions offers significant practical value.
Language: Английский
Citations
2
2022 4th Asia Energy and Electrical Engineering Symposium (AEEES), Journal Year: 2024, Volume and Issue: 225, P. 1044 - 1047
Published: March 28, 2024
Aiming at the problem of transmission line insulator filth type which is difficult to be identified online, this paper proposes identification method based on hyperspectral image, analyzes characteristics and differences different types data, preprocesses spectral data through black-and-white correction multivariate scattering correction, strengthen between types. Based this, an W-SVM proposed realize accurate Finally, it proved by experimental results that accuracy method's reaches 96%, can meet practical needs.
Language: Английский
Citations
0
Forests, Journal Year: 2024, Volume and Issue: 15(7), P. 1139 - 1139
Published: June 29, 2024
Stress wave technology is very suitable for detecting internal defects of standing trees, logs, and wood has gradually become the mainstream in this research field. Usually, 12 sensors are positioned equidistantly around cross-section tree trunks order to obtain enough stress signals. However, arrangement time-consuming laborious, maintaining accuracy imaging under sparse signals a challenging problem. In paper, novel hybrid method based on compressive sensing elliptic interpolation proposed. The spatial structure defective area reconstructed by using advantages signal representation solution waves, healthy space method. Then, feature points selected mixed imaging. comparative experimental results show that overall proposed reaches 89.7%, high-quality effect can be guaranteed when number reduced 10, 8, or even 6.
Language: Английский
Citations
0
Remote Sensing Letters, Journal Year: 2024, Volume and Issue: 15(9), P. 930 - 940
Published: Aug. 19, 2024
Individual tree detection in urban areas using unmanned air vehicles (UAVs) RGB imagery poses challenges due to the diverse shapes and structures of trees complexity forests. The digital surface model (DSM) provides elevation data, fusion UAV with data has emerged as a promising approach for detection. Here, we constructed novel network structure based on faster region-based convolutional neural (Faster R-CNN) detect camphor environments RGB-DSM data. First, an attention module was proposed effectively fuse DSM features by leveraging their complementarity. Second, bidirectional feature pyramid (BiFPN) introduced enhance performance detecting crowns varying sizes. results showed that our could achieved AP 85.7%. Notably, yielded 81.3% green spaces. analysis indicated feasible demonstrated its potential facilitate forestry research applications.
Language: Английский
Citations
0
Computers and Electronics in Agriculture, Journal Year: 2024, Volume and Issue: 227, P. 109498 - 109498
Published: Nov. 1, 2024
Language: Английский
Citations
0