Plants growth is crucial to an agricultural industry, and to the economy of a nation. Consequently, taking care of plants is essential. Like humans, plants are susceptible to several bacterial, fungal, and viral diseases. To avoid the plants from being destroyed, prompt disease detection and treatment are crucial. The goal of this chapter is to introduce plant disease detection and classification using deep learning and transfer learning based pre-trained models. Using images of leaves, the models can identify several plant ailments. The models, namely- Convolutional Neural Network (CNN), MobileNet, and VGG16 are applied for plant disease identification. To train them, the PlantVillage dataset is used, and to enhance the sample size, the dataset is augmented. For experimentation purposes, images of both healthy and damaged plants are taken. Experiment results reveal that VGG16 has outperformed CNN and MobileNet models for the detection of tomato, potato, and apple plant diseases. The accuracy of the VGG16 model is 0.89, 0.92, and 0.95 for the tomato, potato, and apple plant diseases respectively.
TopIntroduction
Being an integral component of our ecosystem, plants are crucial to maintain life on the planet earth. The plants support overall health, well-being of our planet, give us food and air. But much like people and animals, plants can get sick from different pathogens like bacteria, viruses, fungi, and nematodes (Saleem et al., 2019; Kulkarni et al., 2021). These plant diseases can have an adverse effect on the economy and environment, degrade crop quality, and cause large output losses. Effective disease management, which comprises prompt and focused treatments to stop the spread of infections and lessen their negative effects on crop yield, depends on early and accurate diagnosis of plant diseases. Traditional techniques of plant disease identification rely on visual inspection by experienced professionals (Alston et al., 2014) which can be time-consuming, labour-intensive, and subjective that frequently lead to misdiagnosis and delayed treatment. In order to enable prompt intervention and reduce crop losses, there is a rising need for quick, trustworthy, and non-destructive approaches for early plant disease identification (Barbedo, 2018; Eunice et al., 2022).
Recent technological developments have completely changed the way how plant diseases are found, especially in the areas of imaging, sensing, and data analysis. These technological developments have resulted in the creation of numerous cutting-edge methods and instruments for the identification of plant diseases, ranging from molecular techniques and Artificial Intelligence (AI) algorithms to remote sensing and spectroscopy (Narayana et al., 2018). These techniques have potential to provide real-time, precise, and high-throughput plant disease detection, enabling prompt management choices and targeted application of control measures (Talaviya et al., 2020). The objective of this chapter is to present a thorough analysis of the developments in the plant disease detection, with a special emphasis on the different methods and equipments that are emerging in the recent years. This research discusses about the fundamentals, benefits, drawbacks, and applications of several approaches that are used to identify plant diseases, with a focus on their potential for early and non-destructive disease detection (Muhammed, 2022; Javidan et al., 2023). The review also focuses on the difficulties and potential directions in the field of plant disease detection, such as requirements for integrating various strategies, standardizing methods, and creating user-friendly instruments for field/crop use.
There are several worthwhile utilities of plant disease detection and classification models that are stated herewith.
Early Detection and Intervention: The research findings can help to generate early warning systems for plant illnesses, giving farmers the ability to spot diseases in their early stages. This can assist farmers in making early and precise interventions, such spraying pesticides, establishing crop rotation, or modifying irrigation schedules, to stop the spread of illnesses and reduce crop losses (Eunice et al., 2022).
Increased Crop Yields and Food Security: Farmers can use effective disease control techniques and increase crop yields by correctly detecting plant diseases (Li et al., 2021). This can contribute to global food security by ensuring a sufficient and wholesome food supply, especially in areas where agriculture is a major source of income.
Sustainable Agriculture Practices: By limiting the indiscriminate use of pesticides, research can support sustainable agricultural practices (Khan et al., 2021). Farmers can maximize the use of pesticides by detecting diseases early, making focused interventions, lowering costs, and promoting ecologically friendly farming practices.
Cost-Effective Disease Management: Farmers can use cost-effective interventions to optimize their disease management methods by using the research findings. For farmers, this can lead to lower production costs, greater profitability, and higher economic sustainability (Dubey and Jalal, 2013).