Abstract: To improve crop production and protection, and to implement timely targeted pesticide applications, reducing input costs and benefiting the environment, an accurate early detection and quantification of damage caused by crop insect pests in plants is required. Traditional methods such as plant-flapping method which investigates the population of insect pests by macroscopic observation with the tracking down rate between 30% and 70% are most common but subject to bias and can be inaccurate. These imprecise and inaccurate detection and damage evaluation data, however, may cause costly errors to variable-rate spraying in precision agriculture. This paper provides an overview of the recent literatures on machine-based technologies for detecting and monitoring field crop insect pests. Techniques which have been used in detecting and monitoring insect pests include methods of acoustic detection, radar observation and spectral scanning. Some of the main constraints of these progress and solutions where rapid advances seem possible in the machine-based technologies for detecting insect pests of crops are discussed. As for the difficulties in the machine-based technologies for detecting crop insect-pests, such as field conditions complicated variables, injured position uncertainty, and many interference factors, possible approaches are outlined, including that future research should focus on combined detection methods under field conditions.

Key words: Crop pest, remote sensing, acoustic detection, radar observation, video detection, spectral scanning