Effectiveness of Probabilistic Image Sampling Techniques to Identify Hoax-related Images in Indonesia
C. W. D. Lumoindong1, M. A. Aryadi2, I. T. Wilyani3, A. Suhartomo4
1C. W. D. Lumoindong, Department of Electrical Engineering, President University, Cikarang, Indonesia.
2M. A. Aryadi, Department of Electrical Engineering, President University, Cikarang, Indonesia.
3I. T. Wilyani, Department of Electrical Engineering, President University, Cikarang, Indonesia.
4A. Suhartomo, Department of Electrical Engineering, President University, Cikarang, Indonesia.
Manuscript received on 09 January 2020 | Revised Manuscript received on 05 February 2020 | Manuscript Published on 20 February 2020 | PP: 125-131 | Volume-9 Issue-3S January 2020 | Retrieval Number: C10290193S20/2020©BEIESP | DOI: 10.35940/ijitee.C1029.0193S20
Open Access | Editorial and Publishing Policies | Cite | Zenodo | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open-access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Hoaxes are very common among Indonesians. The tendency of most Indonesians to believe everything they saw or heard and the rapid spread of information with questionable credibility in social media contribute to the quick growth of hoaxes. These hoaxes varies from the ‘light’ hoaxes such as April Fools pranks which are taken seriously to some ‘heavier’ ones such as political hoaxes. Fortunately, there are a lot of websites offering hoax identification services. But, most Indonesians would rather holding on the term ‘no picture means hoax’ than checking any kinds of information they received on those websites. As image editing software progressed forward, this old term is not really helpful. Forged images are easily made and spread through social media, and only few Indonesians know how to distinguish between real images and forged images. This research will focus on comparing the probabilistic image sampling techniques in order to combat hoaxes spreading through social media. Before being identified, several images (both forged and real) alongside some opinion-based questions regarding hoax-related imagery will be presented in a form of a survey to 167 respondents, in which most of respondents failed to identify the forged images. The success of the probabilistic image sampling technique will be based on the detection test score of each sampling techniques and their suitability with current situation in Indonesia.
Keywords: Hoax, Probabilistic; Image Processing; Forgery.
Scope of the Article: Image Security