A key component of global protein and oil supply, soybeans are essential for meeting the rising demand for food in the world’s population. Recognized as one of the most vital crops globally, soybean cultivation demands enhanced disease detection and classification methods to mitigate production losses. This literature survey investigates studies spanning from 2018 to 2023, focusing on advancements in soybean disease detection methodologies. We analyze various architectures and methodologies employed in various studies, ranging from traditional machine learning to advanced deep learning algorithms. Through this study, we compare the performance of different models, highlighting their strengths, weaknesses, and potential applications in soybean disease detection. Additionally, we provide insights into the characteristics of common soybean diseases and discuss various approaches for preprocessing soybean images, including data augmentation, segmentation, and feature extraction. Furthermore, we explore the landscape of available soybean disease datasets, noting a reliance on self-acquired datasets for research purposes. Moreover, we propose a new model based on research gaps identified within the literature to address the existing challenges and improve the efficacy of soybean disease detection systems.