EUVIMEDEUVIMED entdecken
Crossref

Advancements In Deep Learning Techniques For ImageBased Detection Of Diseases In Leaves Of Potato: A Review

Hima Gandhi, Madan Lal, Kanwal Preet Singh Attwal

IOSR Journal of Computer Engineering · 2026-01 · 28 (1) · 62-71

DOI: 10.9790/0661-2801016271 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IOSR Journal of Computer Engineering, 2026-01. Verfasst von Hima Gandhi und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This review article analyzes the significant role of deep learning in the automated classification and detection of potato leaf diseases, as they prominently threaten global agricultural production. Traditional approaches to identify diseases typically rely on manual inspection processes that are labor-intensive, time-consuming, and prone to human-induced error, leading to variability and producing inconsistent outcomes. In contrast, deep learning frameworks utilize convolutional neural networks (CNNs), leveraging transfer learning techniques to enhance diagnostic accuracy while reducing computational overhead. By using an extensive dataset of annotated potato leaf images, several studies report that deep learning models can effectively discriminate between healthy and diseased foliage, focusing on prevalent conditions such as Potato Late Blight and Potato Early Blight. This study highlights that deep learning based approaches demonstrate promising performance primarily under controlled and semi-controlled conditions when compared to traditional machine learning methods, while robustness across diverse real-world environments remains an active area of research. Furthermore, the article examines the practical challenges inherent in real-life agricultural environments that include variations in illumination and complex backgrounds that may hinder the process of accurate disease symptom recognition. Moreover, while existing studies evaluate pre-trained and hybrid deep learning models for plant leaf disease detection, these evaluations are often conducted on different crops, datasets, and experimental settings. This highlights a research gap in the systematic and consistent comparison of multiple pre-trained deep learning models for potato leaf disease detection under unified evaluation frameworks, particularly in the context of real-world agricultural conditions. Ultimately, the review aims to expand its scope to ensure broader accessibility to reliable diagnostic tools, thereby enabling farmers and agricultural stakeholders with effective resources to manage potato leaf diseases. By supporting the adoption of such innovative technologies, the study contributes to ongoing efforts aimed at improving agricultural resilience and productivity in the context of global challenges.

Zitieren

APA

Gandhi, H., Lal, M., & Attwal, K. P. S. (2026). Advancements In Deep Learning Techniques For ImageBased Detection Of Diseases In Leaves Of Potato: A Review. IOSR Journal of Computer Engineering, 28(1), 62-71. https://doi.org/10.9790/0661-2801016271

Vancouver

Gandhi H, Lal M, Attwal KPS. Advancements In Deep Learning Techniques For ImageBased Detection Of Diseases In Leaves Of Potato: A Review. IOSR Journal of Computer Engineering. 2026;28(1):62-71. doi:10.9790/0661-2801016271

BibTeX
@article{Gandhi2026,
  author = {Gandhi, Hima and Lal, Madan and Attwal, Kanwal Preet Singh},
  title = {Advancements In Deep Learning Techniques For ImageBased Detection Of Diseases In Leaves Of Potato: A Review},
  journal = {IOSR Journal of Computer Engineering},
  year = {2026},
  volume = {28},
  number = {1},
  pages = {62--71},
  doi = {10.9790/0661-2801016271}
}
Kostenlos starten →