Tree Crop Yield Estimation and Prediction Using Remote Sensing and Machine Learning: A Systematic Review

Smart Agricultural Technology, Journal Year: 2024, Volume and Issue: 9, P. 100556 - 100556

Published: Sept. 1, 2024

Yield prediction has long been a valuable tool for farmers seeking to enhance crop production. Among the many ways predict yield, integration of machine learning (ML) techniques is becoming more common refining methodologies. This study highlights current landscape remote sensing and ML employed in predicting tree yield while also identifying critical gaps areas further exploration. Studies with limited datasets training often use simpler models such as linear regression, studies larger complex models, including deep learning, ensemble methods, hyperparameter tuning; these cases, performance evaluation tends be sophisticated. using demonstrated accuracy levels ranging from 50% 99%. smaller consistently demonstrate higher rates. While can prediction, their effectiveness depends on strategic data collection multi-factor multi-method approach. Integration various sources, weather, soil, plant data, could model resilience applicability. Enhancing research this field achieved through overcoming challenges accurate fostering development open datasets. comprehensive analysis lays groundwork future endeavors aimed at advancing application accurately yield.

Language: Английский

---

The promise of explainable deep learning for omics data analysis: Adding new discovery tools to AI

New Biotechnology, Journal Year: 2023, Volume and Issue: 77, P. 1 - 11

Published: June 16, 2023

Deep learning has already revolutionised the way a wide range of data is processed in many areas daily life. The ability to learn abstractions and relationships from heterogeneous provided impressively accurate prediction classification tools handle increasingly big datasets. This significant impact on growing wealth omics datasets, with unprecedented opportunity for better understanding complexity living organisms. While this revolution transforming these are analyzed, explainable deep emerging as an additional tool potential change biological interpreted. Explainability addresses critical issues such transparency, so important when computational introduced especially clinical environments. Moreover, it empowers artificial intelligence capability provide new insights into input data, thus adding element discovery powerful resources. In review, we overview transformative effects having multiple sectors, ranging genome engineering genomics, radiomics drug design trials. We offer perspective life scientists, understand tools, motivation implement them their research, by suggesting resources they can use move first steps field.

Language: Английский

---

Principles, developments, and applications of spatially resolved spectroscopy in agriculture: a review

Frontiers in Plant Science, Journal Year: 2024, Volume and Issue: 14

Published: Jan. 10, 2024

Agriculture is the primary source of human survival, which provides most basic living and survival conditions for beings. As standards continue to improve, people are also paying more attention quality safety agricultural products. Therefore, detection product very necessary. In past decades, spectroscopy technique has been widely used because its excellent results in detection. However, traditional spectral inspection methods cannot accurately describe internal information With continuous research development optical properties, it found that an object can be better reflected by separating properties light, such as absorption scattering properties. recent years, spatially resolved increasingly field due simple compositional structure, low-value cost, ease operation, efficient speed, outstanding ability obtain about products at different depths. It separate based on transmission equation optics, allows accurate This review focuses principles spectroscopy, equipment, analytical methods, specific applications Additionally, direct analysis reported this paper.

Language: Английский

---

Agriculture 5.0 and Explainable AI for Smart Agriculture: A Scoping Review

Emerging Science Journal, Journal Year: 2024, Volume and Issue: 8(2), P. 744 - 760

Published: April 1, 2024

The visionary paradigm of Agriculture 5.0 integrates Industry 4.0 principles into agricultural practices. Our scoping review explores the landscape 5.0, emphasizing pivotal role Explainable AI (XAI) in shaping this domain. Guided by Preferred Reporting Items for Systematic Review and Meta-Analysis Scoping Review, we rigorously analyzed 84 articles published from 2018 to September 2023. findings highlight XAI’s potential within recognizing its influence on intelligent farming. We propose a conceptual framework integrating XAI, impact model transparency user trust. Despite transformative applications, existing literature often lacks XAI discussions. objective is bridge gap provide reference academics, practitioners, policymakers, educators field smart agriculture that both environmentally friendly technologically advanced. Doi: 10.28991/ESJ-2024-08-02-024 Full Text: PDF

Language: Английский

---

Improving spatial transferability of deep learning models for small-field crop yield prediction

ISPRS Open Journal of Photogrammetry and Remote Sensing, Journal Year: 2024, Volume and Issue: 12, P. 100064 - 100064

Published: April 1, 2024

Predicting crop yield using deep learning (DL) and remote sensing is a promising technique in agriculture. In smallholder agriculture (< 2 ha), where 84% of the farms operate globally, it crucial to build model that can be useful across several fields (high spatial transferability). However, enhancing transferability small-scale setting faces significant challenges, including autocorrelation, heterogeneity scale dependence dynamics, as well need address limited data points. This study aimed test hypothesis cross validation (SCV) more suitable practice than random (RCV) enhance for prediction farming setting. We compared performances DL models predict settings three types two architectures based on RCV with without overlapping samples SCV. Notably, we conducted performance tests external, equally sized instead field used training. high resolution RGB imagery taken drone input. Our results show SCV outperformed those when were tested external (on average r = 0.37 SCV, 0.18 overlap 0.07 without), even though showed substantially lower (CV) (r w/o 0.73 0.98/0.73, respectively). The suggest leads over-optimism by overfitting structure remembering image-specific information (so called memorization). offers first empirical evidence preferable small making transferable.

Language: Английский

---

Exploring the decision-making process of ensemble learning algorithms in landslide susceptibility mapping: Insights from local and global explainable AI analyses

Advances in Space Research, Journal Year: 2024, Volume and Issue: 74(8), P. 3765 - 3785

Published: July 6, 2024

Language: Английский

---

Comparative Result Analysis of Cauliflower Disease Classification Based on Deep Learning Approach VGG16, Inception v3, ResNet, and a Custom CNN Model

Hybrid Advances, Journal Year: 2025, Volume and Issue: unknown, P. 100440 - 100440

Published: March 1, 2025

Language: Английский

---

Counterfactual Based Approaches for Feature Attributions of Stress Factors Affecting Rice Yield

Engineering Reports, Journal Year: 2025, Volume and Issue: 7(1)

Published: Jan. 1, 2025

ABSTRACT Agriculture is a crucial sector in many countries, particularly India, where it significantly influences the economy, food supply, and rural livelihoods. The increased integration of Deep Learning (DL) Machine (ML) into agriculture has enabled substantial advancements predicting crop yields analyzing factors affecting them. counterfactual reasoning framework DICE outperforms LIME offering finer insights feature importance relative impact different on yield prediction. provided clearest causal insights, demonstrating how adjustments to attributes like sandy alfisols surface texture could lead significant changes by water retention nutrient availability. SHAP ranked features phosphate potash based their average across dataset, global view influential but lacking in‐depth understanding. localized immediate influences, such as rainfall nitrogen content, although fell short revealing broader interactions essential for targeted agricultural interventions. findings highlight significance explanations ML models, they provide robust understanding relationships, going beyond correlation‐based attributions. study provides understandable practical allowing focused actions enhance productivity adaptability agriculture. By improving interpretability machine learning research ultimately supports creation predictive systems that strengthen sustainable practices economic development within industry.

Language: Английский

---

High-performance flexible wearable electronics for sheep physiological information wireless sensing and health assessment

Computers and Electronics in Agriculture, Journal Year: 2025, Volume and Issue: 231, P. 109917 - 109917

Published: Jan. 22, 2025

Language: Английский

---

Comprehensive Analysis of a YOLO-based Deep Learning Model for Cotton Plant Leaf Disease Detection

Engineering Technology & Applied Science Research, Journal Year: 2025, Volume and Issue: 15(1), P. 19947 - 19952

Published: Feb. 2, 2025

Diagnosis of cotton plant diseases is essential to maintain agricultural sustainability and output. This study proposes a YOLO-based deep learning model for leaf disease detection maximize accuracy. method ensures comprehensive evaluation health by combining various image processing techniques, improving the accuracy identification. provides viable path improve crop monitoring management in farming systems emphasizes importance utilizing cutting-edge techniques activities. ROC curve performance classification metrics were better YOLOv5 than VGG16 ResNet50, as it had highest F1 score (99.21%), recall, precision. Consistent tests was demonstrated all models, which showed balanced precision, scores. ResNet50 marginally outperformed terms true positive rates, (98.88% vs. 98.65%), More sophisticated such higher efficiency VGG16, makes them more appropriate applications demanding low false rates high The proposed improves identification, ensuring thorough assessment using techniques. results show that approach quite successful correctly detecting classifying variety affect plants.

Language: Английский

---