Answers are grounded in this paper and cite their sources.
Review demonstrates AI-enhanced convolutional neural networks improving rhizome rot disease detection in turmeric and ginger, suggesting better diagnostics.
Rhizome rot infection in Curcuma (turmeric) and Zingiber (ginger) species has become a significant threat to crop yield and quality. The pathogenic infections in these crops are predominantly caused by soil‐borne fungal species. As these crops propagate through subterranean rhizomes, the rot infection remains undetected and untreated until harvest, resulting in substantial productivity loss. Conventional methods of detection have failed to identify infections at an early stage, as symptoms tend to manifest long after infection. Immunosensors and laboratory‐based methods are accurate, yet expensive, labourious, less accessible and time‐consuming. For early and accurate diagnosis, an approach for smart innovative detection, such as integration of laboratory‐based methods with emerging artificial intelligence (AI), can reduce the emerging challenges. The present review explores the application of different convolutional neural network (CNN) models for image‐based plant disease detection and suggests that the GoogLeNet‐Inception‐v3 architecture is best suited for constructing a more accurate CNN model for detecting rhizome rot disease. It also suggests using augmented images as input to measure the accurate test performance. Furthermore, to overcome the issue of cross‐reactivity with other infections in the detection process, a new integrated CNN model has been proposed in the present review. The proposed CNN model will include immunosensor‐based data (bands) and healthy and infected leaf images as input data. This approach promotes digital‐based smart diagnosis for early and accurate detection of rhizome rot, thus enhancing the sustainable management of rhizome rot infection and increasing the productivity of these species.
No takes yet. Share an insight, caveat, or question.
Hota et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: