Optimasi SVM untuk Klasifikasi Warna Investigasi Terhadap Pengaruh Fungsi Kernel dan Penyetelan Parameter

Abstract

Color plays a crucial role in visual applications such as object recognition, image processing, computer vision, and computer graphics. Support Vector Machine (SVM) algorithms have gained attention for color classification due to their ability to handle complex data. SVM, a machine learning algorithm for classification and regression, aims to find optimal decision boundaries. In color classification using SVM, color data is represented by feature vectors, and SVM learns patterns to classify colors accurately. The SVM algorithm demonstrates a high accuracy rate, with an average accuracy of approximately 85% in color detection. This indicates the SVM's ability to effectively separate and classify colors with precision. SVM is proven to be effective in handling non-linear color data by utilizing kernel functions to transform the feature space into higher dimensions, enabling accurate classification of complex color data. The outstanding performance of the SVM algorithm in color detection presents vast potential applications in color recognition, image processing, computer vision, and computer graphics. SVM offers accurate and reliable solutions for object classification based on color characteristics in various contexts.