Whether caused by a fungus, bacteria, or virus, plant diseases can cause significant economic damage to growers. The damage can be direct such as yield loss or indirect through product quality downgrade. As a result, a substantial amount of the production costs is spent on disease inspection, prevention, and fighting the disease. This burdens the farmers financially and increases agriculture’s environmental footprint while also putting the grower’s health at risk by increasing their exposure to harmful chemicals. Modern AI technologies and computer vision hold the promise to automate disease detection through the use of cameras. Agricultural machinery can then use this information to transform blanket applications into targeted ones, meaning that only the diseased parcel of the field/ plant spot is sprayed.

Additionally, this thorough monitoring allows detailed assessing of the disease pressure providing further insights about the optimal course of action. Disease detection models for various plants and diseases have already been developed. Examples are Apple scab and black rot for apples, early blight and Cercospora for Celery, Leaf blight, and Esca for grapes. Most detection models achieved promising results that, on some occasions, reached up to 100%. However, to develop a commercial disease detection solution, more data are needed to create a robust, field invariant disease detection system. The technology and hardware are already available; however, what researchers lack are field images acquired under various illumination and weather conditions, plant and disease growth stages, and different varieties to create a robust and accurate system.