Bioscience Method 2024, Vol.15, No.2, 58-65 http://bioscipublisher.com/index.php/bm 59 2020). Transfer learning approaches using deep learning models like VGG-16 and ResNet have been applied to sugarcane disease classification, further illustrating the potential of these technologies to revolutionize disease detection in agriculture (Daphal and Koli, 2021). Moreover, the development of mobile applications using deep learning architectures like Faster Region-based Convolutional Neural Network (Faster RCNN) has made it possible for farmers to detect diseases in sugarcane crops quickly and efficiently (Murugeswari et al., 2022). These technological innovations represent a significant step towards more sustainable and effective disease management in sugarcane cultivation, which is crucial for maintaining the crop's global economic importance. In summary, the need for technological innovation in the detection and management of sugarcane diseases is clear. The integration of advanced machine learning techniques, remote sensing, and IoT into agricultural practices offers promising solutions to improve disease management and ensure the sustainability of sugarcane production worldwide. 1 Innovations in Disease Detection 1.1 Machine vision and image processing Recent advancements in machine vision and image processing have significantly improved the detection of diseases in sugarcane crops. Support vector machines (SVM) have been utilized to detect sugarcane borer diseases with a high accuracy rate, demonstrating the effectiveness of machine vision technology when combined with threshold segmentation and image processing techniques. Similarly, image processing based disease detection for sugarcane leaves has been implemented, focusing on major diseases such as red rot, mosaic, and leaf scald. The use of computer vision techniques has shown promise in changing the agricultural landscape by enabling automatic disease detection. Furthermore, deep learning frameworks have been proposed to detect diseased sugarcane plants by analyzing leaves, stems, and color, with models like Inception v3, VGG-16, and VGG-19 showing high accuracy in disease identification (Srivastava et al., 2020). The application of transfer learning approaches, such as VGG-16 net and ResNet, has also been explored for sugarcane foliar disease classification, yielding promising results even with limited datasets (Daphal and Koli, 2021). Additionally, the use of Convolutional Neural Networks (CNNs) has been proposed for the automated recognition of sugarcane diseases, further illustrating the potential of machine vision and deep learning in disease detection (Kotekan et al., 2023). 1.2 Molecular diagnostics Molecular diagnostics have emerged as a powerful tool for the management of major diseases in sugarcane. The development of specific and sensitive diagnostic tools using PCR-based detection methods has facilitated the timely detection of pathogens, which is crucial for disease management. These molecular techniques have been applied to a range of sugarcane diseases, including red rot, smut, yellow leaf syndrome, and others, highlighting the need for highly sensitive, specific, and cost-effective detection tools for large-scale applications. 1.3 Remote sensing and UAV technologies Remote sensing and UAV (Unmanned Aerial Vehicle) technologies have been recognized as innovative approaches for plant disease detection. These techniques, coupled with spectroscopy-based methods, allow for the rapid preliminary identification of primary infections and high spatialization of diagnostic results. Novel sensors based on the analysis of host responses, such as differential mobility spectrometers and lateral flow devices, can deliver instantaneous results and effectively detect early infections directly in the field. Biosensors based on phage display and biophotonics have also been developed to provide instantaneous infection detection, which can be integrated with other systems for enhanced disease management. Additionally, the integration of IoT (Internet of Things) in agriculture has led to the development of automated systems that can identify diseases in sugarcane leaves with high accuracy, demonstrating the potential of technology-driven solutions to improve agricultural production and minimize risks (Thangadurai et al., 2020).
RkJQdWJsaXNoZXIy MjQ4ODY0NQ==