Image Processing Algorithm for Estimate the Area of Tungro Disease Attacks on Rice Through Aerial Photography Approach

Abstract

Tungro is one cause of crop failure. The calculation of the extent of the attack is very important to determine how big the attack on a land and to predict the results to be gained. Aerial photographs are one of the technologies that can be used to predict the extent of tungro disease. The development of image processing algorithms should be developed to predict the extent of attacks with aerial photographs.The purpose of this research were to develop an application that can be used to predict the extent of tungro disease on rice. To develop the program of estimate attack area use software Adobe Photoshop CS 6 and Matlab R2013a. This research consists of several steps, namely image acquisition, preprocessing, thresholding, image morphology operation, pixel calculation, calculation of the area of attack.The algorithm of this study were : image acquisition, color normalization, mosaicking, resize, scale calculation, land pixel calculation, pixel attack calculation, calculation of attack area and percentage of attack. T test is performed on data of percentage estimation of attack and manual calculation using planimeter. From the T test results obtained that both data is the same. Based on the above, it can be concluded that the algorithm of image processing estimation to estimate the extent of tungro disease attack is image acquisition, color normalization, mosaicking, resize, scale calculation, land pixel calculation, pixel counting attack and percentage of attack calculation.