Penghitungan Pohon Mangrove dengan Mengidentifikasi Pohon Menggunakan Metode Region-Based Convolutional Neural Network

Abstract

In this study, identification of mangrove trees was carried out and counting the number of trees using the Region-based Convolutional Neural Network (R-CNN) method. Calculation of mangrove trees manually can take a long time and there are limitations that result in an error in the number of existing mangrove trees. Luckily it saves time in counting the number of mangrove trees in a field, so we need an application that can count the number of mangrove trees automatically. The data taken came from the coastal areas of North Sumatra with input in the form of photos in .jpg or .jpeg format with the results taken from the top of the Mangrove tree with a drone. The picture was taken with a perspective perpendicular to the Mangrove tree with the camera height when shooting ± 30 meters from the ground. The results of the experiment by first conducting image training of 280 image data and testing as many as 20 image data with dimensions of 4000 x 3000 pixels, it can be concluded that the application can perform pixel readings of training images with maximum network parameters error of 0.01 and a learning rate of 0.05. The application can detect and calculate the number of mangrove trees with an accuracy value of 90.85%.