We are empowering communication between farms and farmers which are essential in saving them from crop losses and significantly increasing their farming yields.
We are developing a highly efficient framework to empower farmers with pest, disease and weed infestation prediction so that they can take precautionary measures well in advance to avoid crop losses. We collect unique data from their crops using UAVs equipped with Hyperspectral Sensors. These sensors pick up minuscule physiological activities occurring in the crops which are enabled by the data acquisition in the electromagnetic spectrum of 400-1000 nm with more than 100 narrow-bands. We then analyze this data in a co-relation with a spectral database that we are building steadily in the experimental farms of 110 Indian Council for Agriculture Research (ICAR) Institutes. We apply machine learning algorithms to predict the diseases and pest infestation in crops.
This can be understood with a scientific example: When a disease like bacterial blight infects a plant it is caused by a pathogen called Xanthomonas axonopodis pv. punicae. This pathogen releases a number of effector proteins including TAL effectors into the plant through their secretion system. This effector protein causes some biochemical change in the plants and leaves. With human eyes, these changes become visible only when the water absorbing red-brown spots become visible on leaves. We enable the identification of this pathogen at the onset of the infection. Hyperspectral Cameras picks up the humanly invisible color changes occurring in the leaves due to biochemical changes induced by Xanthomonas. Hence, weidentify the disease infestation before a major damage is caused to the plant and becomes visible to naked human eye.