Performance evaluation of the deep learning system for weed recognization

Abstract

Numerous approaches based on machine learning have emerged in recent years to enhance crop protection efficiency. One example is the utilization of deep neural networks (DNNs) to differentiate between various weed types in actual events scenarios. Nevertheless, these methods often need substantial input from experts who work iteratively to design the robust deep learning system. To simplify such process and conserve resources, researchers have explored a fresh method known as automated deep learning our technology’s recognization of weeds through the use of machine learning was evaluated using plant seedlings and weed collections from plants dataset to address a issue of weed recognization. The study compared various configurations, including plant segmentation, using a collection of classifiers in place of Softmax, and training with datasets that contain noise. The findings indicated ensuring performance, with F1-scores of 93.1% and 90.2% based on the dataset utilised. These results align together with automated machine learning (AutoML-linked) studies, while fall short of manually fine-tuned deep-learning-based systems created through human specialists. To conclude, exploring the potential of combining manual expert work and automated deep learning could be a promising direction for enhancing efficiency in plant defence.