Utilizing Ant Colony Optimization With Deep Learning For Wheat Disease Detection

Wheat is a major global food crop that provides over 20% of the calories for humans worldwide. One of the primary staples for a majority of the world's population, making sure that it remains healthy and productive is invaluable. The awareness and early identification of wheat diseases are important for preserving crop yield losses, and to curb pathogen dissemination. By identifying early on, it helps not just in saving this key food source but also to minimize the risk of broad scale outbreaks which may threaten food supply chains. The use of extensive chemical treatments are greatly reduced through effective disease management, hence reducing the likelihood that any harmful chemicals will end up in products made from crops and enter our food chain. In addition, the use of disease detection systems provides resistant measures and innovative methods that allow for more sustainable agricultural practices by increasing resilience to a greater threat. In this paper, we have applied Ant Colony Optimization (ACO) to find out the best hyper-parameters for VGG19 and DenseNet121 models. Instead, we have used ACO that has provided us with creating these models in a super-performing method which can take the shape of precise and efficient training outcomes. It helps increase the accuracy of disease detection along with performance improvements in overall models to foster more accurate and scalable solutions for agriculture technology. With these developments, we hope to contribute towards the larger objective of sustainable and resilient agricultural operations.