Simulation / Modeling / Design

Startup Uses AI to Identify Crop Diseases With Superb Accuracy

Saillog, an Israeli-based startup, developed a mobile application that leverages deep learning to identify over 500 diseases and pest infestations affecting farmers crops. The app can also notify users about which crop diseases and pests have been detected close to their farms.
Using NVIDIA TITAN X GPUs with the cuDNN-accelerated TensorFlow deep learning framework, the startup trained their neural network on hundreds of expert-labeled RGB images of diseased and pest infested crops. Once trained, the mobile application leverages the power of NVIDIA Tesla V100s on the Amazon Web Services for inference.
People anywhere in the world can now upload a picture of their crop and within seconds receive accurate identifications. The app, Agrio, works on both iOS and Android and is available in dozens of languages.
The company says they developed their deep learning system to address a shortage of agriculture professionals who can aid farmers with problems that affect production.

“About 30% of the world’s agricultural yield is lost due to improper management of crops, such as incorrect scanning, monitoring, and treating. These losses occur during the crop growth stage, where disease and pest prevention is critical for optimal output,” Nessi Benishti, Founder, and CEO of Saillog wrote in a blog post. “Saillog utilizes computer vision and artificial intelligence to provide a complete solution for crop disease and pest management,” he added.
The algorithm identifies problems better than experts and is backed by laboratory tests, the startup says.
The company also mentioned they are working on a deep learning robot based on the NVIDIA Jetson platform that could be implemented as an autonomous monitoring solution for farmers.
Read more >

Discuss (0)