Remote sensing for precision weed management

Abstract Herbicides are important for controlling and avoiding the impact of weeds in reducing crop productivity. Traditionally, herbicides are applied uniformly over crop fields. With the advancement of precision agriculture, it is now possible to apply herbicides site‐specifically with precision operations over the crop fields. Precision application of herbicides is crucial to ensure efficient crop productivity. To control and even remove weed interference in crop fields, traditionally, the weeds in crop fields were visually identified and removed manually or killed by chemicals. However, in commercial crop production, accurate applications of herbicides to the target weeds in the crop fields has been a challenge either from the sprayers on aircraft (manned/unmanned) or ground‐based systems. In order to implement precision application of herbicides over crop fields, remote sensing can be effective in identifying the weed targets and assessing the effect of the applied herbicide and its drift. Satellite remote sensing was developed for weed mapping in the 1990s. At the beginning of the 21st century, airborne remote sensing was identified as a promising technique to offer a solution for rapidly identifying and mapping weeds in crop fields for precision weed management. In the early 2010s, unmanned aerial vehicles (UAVs) were developed and used for weed management. Since then, there has been growing interest in using UAVs as a remote sensing platform for precision weed management in crop field. The applications include weed mapping over crop fields, generating prescriptions for precision herbicide application, assessment of crop injury from aerially applied off‐target glyphosate, identification and differentiation of glyphosate‐resistant weeds from glyphosate‐susceptible weeds, and assessment of crop injury from field‐applied dicamba. With the current advancements made in developing site‐specific digital agricultural technology, it is possible to target weeds of interest for precision spot‐spraying application. However, large‐scale datasets play a vital role in developing computer vision‐based system that can accurately target weeds in any environment. The current challenge that the agriculture industry faces is not a lack of expert engineers or scientists to develop site‐specific technology, but rather, adaptability. For instance, a drone‐based spot‐spraying approach developed in the state of California should be able to generalize and adapt to the environment in Michigan or other states. Only then can experts from multiple fields scale up sustainable technology, rather than develop it just for one state or location. Therefore, this study will help readers gain in‐depth understanding of the approaches needed to develop a robust precision application technology for site‐specific weed management. © 2025 Society of Chemical Industry.