Early detection of sugarcane smut using hyperspectral imaging and deep learning techniques

Sugarcane, a member of the grass family known for its high sucrose content, is a valuable source for producing white sugar and by-products such as molasses and bagasse. However, diseases affecting sugarcane crops can render them unproductive, making rapid disease detection essential. One of the most significant diseases is sugarcane smut, caused by the fungus Sporisorium scitamineum. This disease poses a serious threat to sugarcane-growing regions in Iran and worldwide, with potential yield reductions ranging from ۳۰% to ۱۰۰%. Early detection and prompt response to sugarcane disease outbreaks are critical for achieving sustainable agricultural practices, enabling farmers and agricultural professionals to take timely measures to prevent disease spread and minimize yield losses. However, this task is challenging because many sugarcane diseases are difficult to detect in their early stages due to the absence of visible symptoms. Traditional disease detection methods are labor-intensive and time-consuming. With the advent of artificial intelligence (AI), advanced tools are now available to diagnose diseases more accurately and efficiently. This article provides a comprehensive analysis of several key publications on the application of AI, emphasizing new and advanced methods for the early detection of sugarcane smut disease. The study also examines the challenges, issues, and prospects associated with these methods. The findings highlight the importance of AI-based systems, particularly those utilizing image processing and deep learning, which have shown promising results in agriculture. By applying AI algorithms, farmers and agricultural specialists can identify diseases with high accuracy, reduce losses, and optimize the use of agricultural chemicals.

کلیدواژه ها:

Artificial Intelligence (AI) ، Disease outbreak ، Hyperspectral images (HSIS) ، Whip smut