Deep learning networks-based tomato disease and pest detection: a first review of research studies using real field datasets DOI Creative Commons
Mohieddine Jelali

Frontiers in Plant Science, Год журнала: 2024, Номер 15

Опубликована: Окт. 25, 2024

Recent advances in deep neural networks terms of convolutional (CNNs) have enabled researchers to significantly improve the accuracy and speed object recognition systems their application plant disease pest detection diagnosis. This paper presents first comprehensive review analysis learning approaches for tomato plants, using self-collected field-based benchmarking datasets extracted from real agricultural scenarios. The shows that only a few studies available literature used data fields such as PlantDoc dataset. also reveals overoptimistic results huge number PlantVillage dataset collected under (controlled) laboratory conditions. finding is consistent with characteristics dataset, which consists leaf images uniform background. uniformity background facilitates classification, resulting higher performance-metric values models. However, models are not very useful practice, it remains desirable establish large diseases With some self-generated reviewed this paper, high performance above 90% can be achieved by applying different (improved) CNN architectures Faster R-CNN YOLO.

Язык: Английский

Resource constraint crop damage classification using depth channel shuffling DOI
Md. Tanvir Islam,

Safkat Shahrier Swapnil,

Md Mashum Billal

и другие.

Engineering Applications of Artificial Intelligence, Год журнала: 2025, Номер 144, С. 110117 - 110117

Опубликована: Янв. 29, 2025

Язык: Английский

Процитировано

0
Mobile robot for leaf disease detection and precise spraying: Convolutional neural networks integration and path planning DOI Creative Commons

Youssef Bouhaja,

Hatim Bamoumen,

Israe Derdak

и другие.

Scientific African, Год журнала: 2025, Номер unknown, С. e02717 - e02717

Опубликована: Апрель 1, 2025

Язык: Английский

Процитировано

0
Deep learning networks-based tomato disease and pest detection: a first review of research studies using real field datasets DOI Creative Commons
Mohieddine Jelali

Frontiers in Plant Science, Год журнала: 2024, Номер 15

Опубликована: Окт. 25, 2024

Recent advances in deep neural networks terms of convolutional (CNNs) have enabled researchers to significantly improve the accuracy and speed object recognition systems their application plant disease pest detection diagnosis. This paper presents first comprehensive review analysis learning approaches for tomato plants, using self-collected field-based benchmarking datasets extracted from real agricultural scenarios. The shows that only a few studies available literature used data fields such as PlantDoc dataset. also reveals overoptimistic results huge number PlantVillage dataset collected under (controlled) laboratory conditions. finding is consistent with characteristics dataset, which consists leaf images uniform background. uniformity background facilitates classification, resulting higher performance-metric values models. However, models are not very useful practice, it remains desirable establish large diseases With some self-generated reviewed this paper, high performance above 90% can be achieved by applying different (improved) CNN architectures Faster R-CNN YOLO.

Язык: Английский

Процитировано

2