Researchers Develop AI Tool to Detect Tomato Diseases in the Field

Academics from Charles Darwin University (CDU), the University of Peradeniya (UoP), and partners developed a nearly 9,000-image tomato leaf dataset to improve AI-powered crop disease detection under real-world field conditions. The Sri Lankan In-Field Tomato (SLIF-Tomato) dataset contains images of healthy leaves and seven major diseases, providing a foundation for developing faster and more accessible tools for farmers.

The researchers used an AI model called Inverted Residual Convolutional Block Attention Module (IR-CBAM) to help the system focus on important disease features such as color, texture, edges, and affected areas while ignoring background clutter. Study lead author Romiyal George said, “The lightweight AI models developed in this research can enable rapid disease identification using resource-constrained devices such as mobile phones or embedded systems.”

The system achieved more than 99% accuracy in detecting tomato diseases, demonstrating its potential for early diagnosis outside controlled laboratory settings. George said the technology could reduce crop losses, improve treatment decisions, and support more efficient use of agricultural inputs. The researchers plan to test the technology in real-world conditions and expand datasets to include more crops and diseases.

For more information, read the article from Charles Darwin University, Australia.