Deteksi Objek Daun Tebu Dengan Menggunakan Metode Klasifikasi Pada Machine Learning

Abstract

The creation of a machine learning model that combines the Support Vector Machine (SVM) classification algorithm with the Histogram of Oriented Gradient (HOG) feature to detect sugarcane leaf objects. This research utilizes 227 processed images from 829 image data from Kaggle. The data is divided into two classes, namely other (class 0) and sugarcane (class 1). Analysis was performed using Python on the Google Colab platform. The model achieved 96% accuracy, precision 0.96, recall 1.00 and F1-score 0.98. This study provides a solution for automatic detection of tree count objects in Plantation land.