A Self-Construction Automatic Crescent Sighting Detection With Harr - Cascade Classifier Using Adaboost Algoritm and Support Vector Machine
Robiatul Muztaba (a*,b,e), Hakim L. Malasan (b,c,e), Mitra Djamal (d)

a) Doctoral Program in Astronomy, Faculty of Mathematics and Natural Sciences, Institut Teknologi Bandung, Jl. Ganesha No.10, Bandung 40132, Indonesia
b) Departement of Atmospheric & Planetary Sciences, Faculty of Science, Institut Teknologi Sumatera, Lampung 35365, Indonesia
c) Department of Astronomy, Faculty of Mathematics and Natural Sciences, Institut Teknologi Bandung, Bandung 40132, Indonesia
d) Department of Physics, Faculty of Mathematics and Natural Sciences, Institut Teknologi Bandung, Bandung, 40132, Indonesia
e) Observatorium Astronomy ITERA Lampung (OAIL), Institut Teknologi Sumatera, Lampung 35365, Indonesia

Abstract

The automatic detection method using computer vision applied to the crescent moon is a novel concept that will be further developed. This will be highly useful for observers during the process of observing the crescent visibility. This paper proposes a method for high-performance crescent detection based on visual attention mechanism and AdaBoost cascade classifier. Our method constructs the structural Haar features and extracts the features of samples using structural Haar features and trains an AdaBoost cascade classifier. Then we use the visual attention mechanism to extract the target candidate region. At last, we generate detecting sub-windows in the candidate region and discriminate them with the cascade classifier to realize crescent detection. The results obtained have proven excellent detection performance not only when the crescent is visible in front of the camera but also when the crescent is partially covered by clouds. In addition, our approach can be applied during real-time experiences.

Keywords: Automatic Detection- Crescent Moon- Computer Vision

Topic: Industry 4.0