PERANCANGAN SISTEM PENGENALAN BENTUK MUTIARA DARI CITRA TIRAM MUTIARA HASIL RADIOGRAFI SINAR-X DIGITAL
Abstract
Abstrak
Telah dirancang sebuah sistem pengenalan bentuk mutiara dengan memanfaatkan citra digital hasil radiografi sinar-X. Penelitian ini bertujuan untuk membantu mendeteksi ketidakseragaman bentuk mutiara. Teknik pengolahan citra dan algoritma pengenalan pola dengan metode rantai digunakan untuk merepresentasikan kontur citra, Sistem dirancang menggunakan bahasa pemrograman berbasis Borland Delphi 7, dengan alur sistem dimulai dari inputan citra tiram mutiara berformat BMP (grayscale), pengolahan citra, hingga output sistem berupa perbandingan ukuran panjang antara jari-jari yang saling tegak lurus. Jika perbandingannya sama, maka citra mutiara yang dideteksi berbentuk lingkaran. Hasil perancangan sistem yang diperoleh menunjukan bahwa algoritma pengenalan pola mampu mendeteksi bentuk citra mutiara. Hasil deteksi akan semakin baik jika citra hasil radiografi sinar-x yang diinput memiliki kualitas yang baik pula,
Kata kunci: mutiara, pengolahan citra, algoritma pengenalan pola, metode kode rantai, deteksi bentuk.
Abstract
Has designed a software to recognize the form of pearls by utilizing digital image radiographic X-ray results. This study aims to help detect unevenness shape of pearls. Techniques of image processing and pattern recognition algorithms with the chain method is used to represent the image of the contour, the system is designed to use the programming language of Borland Delphi 7, based with the Groove system starting from the input image formats such as BMP Pearl oysters (grayscale), image processing, and output of the system in the form of a comparison length between the fingers perpendicular to each other. If the comparison is the same, then the image of Pearl detected circular. System design results obtained showed that pattern recognition algorithms capable of detecting image forms a Pearl. The results of the detection will be getting better if the image of x-ray radiography results are inputted has a good quality.
Keywords: pearl, image processing, pattern recognition algorithms, chain code method, the detection of the shape.
Downloads
Jl. Adisucipto, Penfui-Kupang, Lasiana, Klp. Lima, Kota Kupang, Nusa Tenggara Timur., Indonesia
This work is licensed under Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0)