Attention-Augmented Multiscale Dilated Deep Neural Networks for Early Detection of Diseases and Nutrient Deficiencies in Crops

Plant diseases and nutrient deficiencies significantly threaten agricultural productivity, particularly in commercial/cash crops such as cotton and palms. These conditions can lead to reduced yields and economic losses for farmers, underscoring the importance of timely and accurate diagnosis. Early detection plays a crucial role in implementing effective management strategies to mitigate these impacts. This study presents a novel classification approach utilizing attention-enhanced multiscale dilated convolutional neural networks (CNNs) for the detection of diseases in cotton and date palm crops. By utilizing multi-scale dilated convolutions, the model effectively captures discriminatory features at various scales, captures features at various resolutions without significantly increasing the parameter count, allowing the network to focus on both fine and coarse details of the disease symptoms. The inclusion of an attention mechanism allows the network to focus on the most relevant features, further enhancing efficacy of the proposed framework. Experimental results demonstrate that the model achieves an accuracy of 96.42% for cotton and 94.93% for palm, outperforming existing techniques. The findings indicate that the proposed approach is highly effective for the early detection of diseases and deficiencies in cash crops and can be beneficial for farmers and agricultural practitioners aiming to improve crop health and optimize yield.