Products from wheat are available almost everywhere. Looking at various products such as bread and other food items, we see that most of them are made with some component of wheat. Therefore, we could see how ubiquitous the presence of wheat is in our nutrition.
When producing wheat, however, there might be various diseases attached to them. There could be diseases such as Stripe Rust that affects the growth of wheat. As a result, the produce for that particular field is reduced to a large extent.
Identifying and combating this task seems to be the primary challenge for farmers. This is because they would have to manually look at all the crops and carefully analyze their health before taking steps to prevent them. As it is manually time consuming to gauze the health of wheat as there are lots of them available in fields, it would be good to use artificial intelligence and deep learning to combat this problem.
With an improvement in technology and computational resources, it has become easier to train and deploy deep learning models, specifically for tasks such as computer vision. Therefore, we could leverage this technology to identify various diseases associated with wheat without manually having to identify them. As a result, this would save a lot of time and effort on the part of the farmers and research scientists in the field of agriculture respectively.
- The computer vision model that was performing the best in classifying the diseases in wheat plants is VGG19.
- The accuracy for classifying the diseases in wheat plants was 95% on the test data (data that the model has not seen before).
- Furthermore, VGG19 was also being deployed in real-time where users can get to upload their own images of wheat and get to know if the wheat plant in the images is healthy or has diseases.
- More high precision images from wheat plants could be collected and analyzed under different climatic and lighting conditions.
- It could be integrated with a camera where it is able to provide regular updates about the condition of wheat plants.