A Method of Constructing Models for Estimating Proportions of Citrus Fruit Size Grade Using Polynomial Regression

Tanimoto, Yuu; Yoshida, Shinichi

doi:10.3390/agronomy14010174

Cited by 3 publications

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

Data Generation Using Pix2pix to Improve Yolo V8 Performance in Uav-Based Yuzu Detection

Zhang,

Tanimoto,

Iwata

et al. 2024

Preprint

View full text Add to dashboard Cite

Data Generation Using Pix2pix to Improve Yolo V8 Performance in Uav-Based Yuzu Detection

Zhang,

Tanimoto,

Iwata

et al. 2024

Preprint

View full text Add to dashboard Cite

Dynamic perspectives into tropical fruit production: a review of modeling techniques

Mancero-Castillo,

Garcia,

Aguirre-Munizaga

et al. 2024

Front. Agron.

View full text Add to dashboard Cite

Modeling the intricate interactions between fruit trees, their environments, soils, and economic factors continues to be a significant challenge in agricultural research globally, requiring a multidisciplinary approach. Despite advances in agricultural technology and algorithms, significant knowledge gaps persist in understanding and modeling these interactions. This review explores basic concepts related to modeling for tropical fruit production. It explains modeling development from sensor technologies, image analysis, databases, and algorithms for decision support systems while considering climate changes or edaphoclimatic limitations. We report the current fruit modeling tendencies showing a significant increase in publications on these topics starting in 2021, driven by the need for sustainable solutions and access to large agricultural databases. This study emphasizes inherent challenges in tropical fruit modeling, such as fruit tree cycles, costly and time-consuming experimentation, and the lack of standardized data. These limitations are evident in tropical fruit, where few models have been reported or validated for cocoa, avocado, durian, dragonfruit, banana, mango, or passion fruit. This study analyzes the classification of the algorithms related to tropical fruit into three main categories: supervised, unsupervised, and reinforcement learning, each with specific applications in agricultural management optimization. Crop classification and yield prediction use supervised models like neural networks and decision trees. Unsupervised models, like K-Means clustering, allow pattern identification without prior labels, which is useful for area segmentation and pest detection. Automation of irrigation and fertilization systems employs reinforcement learning algorithms to maximize efficiency. This multidisciplinary review discusses recent approaches to 1) Modeling Soil health and plant-soil interaction, 2) Yield prediction in tropical fruit orchards, 3) Integrating meteorological models for enhanced tropical fruit production, and 4) Economics of tropical fruit business through modeling. Furthermore, this review illustrates the complexity and multidisciplinary research on models for tropical fruit and platforms using agricultural models. Further opportunities to advance fruit modeling frameworks are indicated, requiring technical knowledge about the fruit crop requirements with user-friendly platforms to collect and access fruit tree data and site-specific agroecological conditions.

show abstract

Object Detection for Yellow Maturing Citrus Fruits from Constrained or Biased UAV Images: Performance Comparison of Various Versions of YOLO Models

Tanimoto,

Zhang,

Yoshida

2024

AgriEngineering

View full text Add to dashboard Cite

Citrus yield estimation using deep learning and unmanned aerial vehicles (UAVs) is an effective method that can potentially achieve high accuracy and labor savings. However, many citrus varieties with different fruit shapes and colors require varietal-specific fruit detection models, making it challenging to acquire a substantial number of images for each variety. Understanding the performance of models on constrained or biased image datasets is crucial for determining methods for improving model performance. In this study, we evaluated the accuracy of the You Only Look Once (YOLO) v8m, YOLOv9c, and YOLOv5mu models using constrained or biased image datasets to obtain fundamental knowledge for estimating the yield from UAV images of yellow maturing citrus (Citrus junos) trees. Our results demonstrate that the YOLOv5mu model performed better than the others based on the constrained 25-image datasets, achieving a higher average precision at an intersection over union of 0.50 (AP@50) (85.1%) than the YOLOv8m (80.3%) and YOLOv9c (81.6%) models in the training dataset. On the other hand, it was revealed that the performance improvement due to data augmentation was high for the YOLOv8m and YOLOv9c models. Moreover, the impact of the bias in the training dataset, such as the light condition and the coloring of the fruit, on the performance of the fruit detection model is demonstrated. These findings provide critical insights for selecting models based on the quantity and quality of the image data collected under actual field conditions.

show abstract

A Method of Constructing Models for Estimating Proportions of Citrus Fruit Size Grade Using Polynomial Regression

Cited by 3 publications

References 28 publications

Data Generation Using Pix2pix to Improve Yolo V8 Performance in Uav-Based Yuzu Detection

Data Generation Using Pix2pix to Improve Yolo V8 Performance in Uav-Based Yuzu Detection

Dynamic perspectives into tropical fruit production: a review of modeling techniques

Object Detection for Yellow Maturing Citrus Fruits from Constrained or Biased UAV Images: Performance Comparison of Various Versions of YOLO Models

Contact Info

Product

Resources

About