Main Article Content
Prediction of agricultural products is needed in terms of planning and decision making as well as in policy making for national food security. One strategic commodity that needs special attention is rice. This study aims to predict rice yields using unmanned aircraft. The results of image acquisition are processed by multi thresholding method to separate leaf objects, rice panicles, and background. Furthermore, the results of sorting objects are used as input in making predictions of rice crop models using artificial neural networks. To compare the results of predictions, we weighed the weight of rice harvest on each block. The results showed that between predictions and actual correlations were very strong with R2 = 0.88, MSE (Mean Square Error) = 0.169 and MAPE (Mean Absolute Percentage Error) values were -0.006. These results indicate that the prediction model of rice yields can be used for estimation purposes.
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.