Crop And Weed Detection System

Machine learning plays a critical role in making crop forecasting decisions, including supporting decisions to differentiate whether the plant is a crop or weed.Weeds are the undesirable plant in agriculture as they compete with desirable crops and use up water, nutrients, and space.Some plants can also enter the machines and prevent the harvest.Therefore, a weed control system is a necessity.In recent years, advancements in computer vision and machine learning have led to the development of various techniques and tools for weed and crop detection.In this study, we conducted a systematic literature review (SLR) to extract and synthesize algorithms and features used in research on crop and weed detection.Based on our search criteria, we collected 567 relevant studies from 6 electronic databases and selected 50 additional studies for further analysis using inclusion and exclusion criteria.We carefully evaluate these studies, identify the methods and features used, and provide recommendations for further research.According to our analysis, the most frequently used algorithms are neural networks.After checking that 50 documents based on machine learning were identified, we conducted further searches in electronic databases to identify deep learning-based studies, retrieved 30 data based on deep learning, and provided data using the Deep learning algorithm.According to this additional analysis, convolutional neural networks (CNN) are the most commonly used deep learning algorithms in these studies; Other deep learning algorithms are long-term memory (LSTM) and deep neural networks (DNN).