Advances in Deep Learning Applications for Plant Disease and Pest Detection: A Review

Remote Sensing, Год журнала: 2025, Номер 17(4), С. 698 - 698

Опубликована: Фев. 18, 2025

Traditional methods for detecting plant diseases and pests are time-consuming, labor-intensive, require specialized skills resources, making them insufficient to meet the demands of modern agricultural development. To address these challenges, deep learning technologies have emerged as a promising solution accurate timely identification pests, thereby reducing crop losses optimizing resource allocation. By leveraging its advantages in image processing, technology has significantly enhanced accuracy disease pest detection identification. This review provides comprehensive overview recent advancements applying algorithms detection. It begins by outlining limitations traditional this domain, followed systematic discussion latest developments various techniques—including classification, object detection, semantic segmentation, change detection—to Additionally, study highlights role large-scale pre-trained models transfer improving scalability across diverse types environmental conditions. Key such enhancing model generalization, addressing small lesion ensuring availability high-quality, training datasets, critically examined. Emerging opportunities monitoring through advanced also emphasized. Deep technology, with powerful capabilities data processing pattern recognition, become pivotal tool promoting sustainable practices, productivity, advancing precision agriculture.

Язык: Английский

---

Transformative Deep Learning Approaches for Enhanced Image Analysis and Processing

Advances in computational intelligence and robotics book series, Год журнала: 2025, Номер unknown, С. 329 - 378

Опубликована: Март 7, 2025

This chapter examines how transformative deep learning is revolutionizing image processing and analysis, especially in the context of complex imaging tasks. Even with major improvements, accuracy efficiency issues are still common. To address these challenges, we discussed different methods that integrate architectures, such as convolutional neural networks (CNNs), RCNN their variants, sophisticated data preprocessing approaches. A thorough analysis model architectures demonstrates significant advantages provides over conventional techniques, improving diagnostic precision effectiveness while facilitating individualized care a variety fields, including remote sensing, self-driving vehicles, medical imaging. In chapter, critically review literature, represent step forward applications for advanced processing, demonstrating its potential to current limitations drive future advancements.

Язык: Английский

---

EfficientNetB3 for High-Performance Insect Identification

Опубликована: Авг. 28, 2024

Язык: Английский

---