ANALISIS PENDETEKSI KECOCOKAN OBJEK PADA CITRA DIGITAL DENGAN METODE ALGORITMA SIFT DAN HISTOGRAM COLOR RGB

Rosidin Al Caruban, Bambang Sugiantoro, Yudi Prayudi

Abstract


Through using tools of image processing on digital images just like gimp and adobe photoshop applications, an image on digital images can be a source of information for anyone who observes it. On one hand, those applications can easily change or manipulate the authenticity of the image. On the other hand, they can be misused to undermine the credibility of the authenticity of the image in various aspects. Thus, they can be considered as a crime. The implementation of the SIFT Algorithm (Scale Invariant feature transform) and RGB color histogram in Matlab can detect object fitness in digital images and perform accurate test.

This study discusses the implementation of getting object fitness on digital image that has been manipulated by SIFT Algorithm method on the Matlab source. It is done by comparing the original image with the manipulated one. The object fitness in digital images can be obtained from a number of key points and other additional parameters through comparing number of pixels on the analyzed image and on the changed histogram in RGB color on each analyzed image.

The digital image forensics which is known as one of the scientific methods commonly used in researches is aimed to obtain evidences or facts in determining the authenticity of the image on digital images. The use of the SIFT algorithm is chosen as an extraction method because it is invariant to scale, rotation, translation, and illumination changes. SIFT is used to obtain characteristics of the pattern of the gained key point. The tested result of the SIFT Algorithm method (Scale Invariant feature transform) is expected to produce a better image analysis.


Full Text:

PDF


DOI: https://doi.org/10.14421/csecurity.2018.1104

Refbacks

  • There are currently no refbacks.


Copyright (c) 2018 Rosidin Al Caruban, Bambang Sugiantoro, Yudi Prayudi

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.