Integrating CNN and KNN for Automated Diagnosis of Stripe Rust in Wheat

In this painting, researchers created and assessed a version to categorize wheat leaf Stripe Rust ailment into ten wonderful Condition Intensity tiers. F1 rankings, consideration, and precision for all training display the right overall performance. The quality precision (97.77%), bear in mind (97.83%), and F1-rating (97.80%) had been proven by way of Condition Intensity nine, indicating the version’s accuracy in figuring out extreme ailment intensities. High overall performance was also verified utilizing the bulk of different conditions, including Condition Intensity 6 (97.58% precision, 97.50% bear in mind, 97.54% F1-rating) and Condition Intensity 7 (97.60% precision, 97.67% recall, 97.63% F1-rating). Even though Condition Intensity 10 had fairly decreased ratings, with 92.81% precision, 92.59% consider, and 92.70% F1-score, the model nonetheless maintained correct accuracy. All lessons were performed with an accuracy of 99%, and the consequences were uniform in terms of support and accuracy. With the most important assist matter (1,431) and a percentage of 0.14, Condition Intensity 9 verified the steadiness of the model whilst implemented to large datasets. With 10,245 photos used to train it, the simulation yielded a general accuracy of 93.02%. Lower depth conditions showed the model’s ability to address fewer, not unusual instances well; as an instance, Condition Intensity eight with a guide be counted of 532 and Condition Intensity 10 with 432 each maintained 99% accuracy. Overall, the consequences show that the model can accurately and precisely categorize numerous ailment depth degrees with sturdy recall, accuracy, F1 ratings, and precision, which makes it useful for actual-world wheat ailment management packages.