Image Based Automated Weather Analysis Using Deep Learning for Decision Support to Remote Farmers

Jeen Retna Kumar R, Berakhah F. Stanley, Subha S., Jim Reeves S., Gold Beulah Patturose J., P. Bini Palas

International Journal of Intelligent Systems and Applications in Engineering · 16 July 2023

In one sentence. CNN plus Bi-GRU classifies weather conditions from on-farm sky images to support farmers' decisions; no quantitative results given.

Abstract

In farming, weather forecasting refers to the prediction of an area's atmospheric state at a specific moment to determine whether it is suitable or unsuitable for agricultural activities. Knowledge about the current weather feed good decision for farmers in crop cultivation. The numerous daily decisions can be better prepared based on weather analysis. These choices include when to fertilize, when to irrigate crops, and when to do fieldwork in the farm. Farmers' choices will determine whether or not their crop is lucrative. A farmer must be mindful of the present climate condition in order to grow a successful crop. In recent years, estimating climate has gained significant research attention. Despite the abundance of data available for weather forecasting, the majority of them require the use of a nonlinear model before forecasting. Due to the advancement in image processing, climate detection from atmosphere images provides much information for identifying weather condition and it helps the farmers choose the appropriate time to operate effectively on a daily basis. This article examines the applicability of the combined convolutional neural network (CNN) and bidirectional gated recurrent unit (Bi-GRU) approach by developing a robust and climate estimation model from real time images acquired from farm atmosphere. The proposed work offers a method for precisely and accurately detecting actual meteorological conditions using computer vision.

Labels

Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.