计算机科学
跳跃式监视
人工智能
聚类分析
过程(计算)
人工神经网络
目标检测
植物病害
对象(语法)
机器学习
网格
模式识别(心理学)
数学
生物技术
操作系统
生物
几何学
作者
Chairma Lakshmi K R,B Praveena,G Sahaana,Nithya Jenev J,T. Gnanasekaran,Mohammed Homod Hashim
标识
DOI:10.1109/icais56108.2023.10073875
摘要
Identifying diseases on crops is an essential but time-consuming process in farming methods. Along with competent labor, it takes a lot of time. The existing systems like K-mean clustering technique and the neural network can detect only 80-90% of illnesses. Using computer vision and machine learning approaches, this research suggests an innovative and effective method for identifying crop pathology. The proposed approach uses the object detection technique's experimental solution, known as YOLO, to find plant disease. YOLO processes leaf images at a real-time rate of 45 frames per second, which is quicker than existing object detection methods. Before processing the image, it divides the image into a number of grid cells. A single neural network can forecast the bounding boxes and class probabilities in a single assessment. With the use of the proposed algorithm, farmers will be able to identify diseases in their early stages, diagnose leaf diseases, and then control the crop as needed to maintain the health and safety of plants.
科研通智能强力驱动
Strongly Powered by AbleSci AI