Crop monitoring and disease detection: Using statistical analysis of sensor data to detect crop diseases and pests

This review paper provides the state of art in using statistical analysis of sensors for accurate and early identification of diseases and pest in crops. Discussed the multiplied applications of hyperspectral, multispectral, thermal and LiDAR approaches in precisions agriculture and highlighting their roles in distinguishing small changes that are Potential stress or infestation. The essence of the review centers on the intricately applied methodological analysis of the sensor data through the application of machine learning algorithms, multivariate analysis of variance for the reduction of the voluminous sensor data into intelligent and manageable patterns, and time series modeling. Discussed the relative advantages and disadvantages of diverse analytical methods, pay attention to the importance of the proper choice of an approach depending on the data type and the nature of a disease/pest. Moreover, some of the main issues in the preprocessing of text, feature engineering, and model evaluation to consider the importance of stable analytical workflows are discussed. Last but not the least, outline possible future directions of development in this swiftly growing field, with a focus on the primary future directions that consist in increasing the use of artificial intelligence and big data analytics for the purposes of more accurate predictions and in development of user-friendly tools for farmers.