Thursday 23 January 2025
Tomato crops are a vital part of global food security, but they’re often threatened by diseases that can significantly impact yields and quality. Detecting these diseases early on is crucial to preventing widespread damage, but it’s a challenging task for farmers and researchers alike.
A team of researchers has been working on developing an AI-powered system that can quickly and accurately identify tomato leaf diseases using images taken with smartphones or digital cameras. Their approach involves training deep learning models on a large dataset of images featuring different types of diseased leaves.
The researchers used two well-established convolutional neural networks (CNNs), VGG19 and Inception v3, to develop their system. They trained the models on a dataset known as Tomato Villages, which consists of over 4,500 images of healthy and diseased tomato leaves.
To improve the accuracy of their system, the researchers used data augmentation techniques such as flipping, rotating, and zooming to create more training examples. They also normalized the pixel values in each image to ensure that the models received consistent input.
The results were impressive: the VGG19 model achieved an accuracy of 93.9% on a separate test set, while the Inception v3 model scored 94.5%. The researchers also found that their system was able to correctly identify diseases such as magnesium deficiency and spotted wilt virus with high accuracy.
One of the key challenges in developing this system was addressing class imbalance, which occurs when some classes have significantly more examples than others. To overcome this issue, the researchers used a technique called data augmentation to generate more training examples for the less-represented classes.
The system developed by the researchers has the potential to be a game-changer for tomato farmers and researchers. By allowing them to quickly and accurately identify diseases in their crops, it could help reduce crop losses and improve yields. The technology could also be adapted for use with other types of crops, making it a valuable tool for agricultural research and development.
In the future, the researchers plan to continue refining their system by exploring new data augmentation techniques and developing more advanced models. They’re also working on integrating their system with existing farm management software to make it easier for farmers to use. With its potential to improve crop yields and reduce disease losses, this AI-powered system could play a vital role in ensuring global food security.
Cite this article: “AI-Powered System Detects Tomato Leaf Diseases with High Accuracy”, The Science Archive, 2025.
Tomato, Leaf Diseases, Ai-Powered System, Deep Learning Models, Convolutional Neural Networks, Data Augmentation, Image Classification, Crop Yields, Food Security, Disease Diagnosis
