Optimization of Underwater Image Objects with Noise Analysis Using a Gaussian Filter Selected Algorithm.

Sukemi, Sukemi (2020) Optimization of Underwater Image Objects with Noise Analysis Using a Gaussian Filter Selected Algorithm. advances on intelligent systems research, Sriwijaya International Conference on Information Technology and Its Applications (SICONIAN 2019), 172. pp. 269-275. ISSN 1951-6851

[thumbnail of Proceeding-Siconian.pdf]
Preview
Text
Proceeding-Siconian.pdf

Download (1MB) | Preview
[thumbnail of Cover Pengelola-Daftar Isi.pdf]
Preview
Text
Cover Pengelola-Daftar Isi.pdf

Download (142kB) | Preview

Abstract

Different types of images when testing will give different results. Between still images and moving images requires speed and accuracy in the computation process. This is due to the availability of time in processing which tends to narrow at any time along with environmental changes. Retesting of processors that have been built through previous research requires the selection of new image data and processes. The selection of the new image refers to a different image and has never been used before. Next, for the new process by applying Gaussian filtering selection (filters block). The results of the first stage in testing of some images obtained that the „bit space adder/sub‟ accuracy value and using filters block for underwater objects image data test was 90.91% in the first cycle. However, when compared to the architecture of least significant bit only obtained an accuracy of 0.01% so that there is a very significant difference in accuracy, which is equal to 90.90%. This result will improve if added using a filter block, the accuracy value rises to 95.75%.

Item Type: Article
Subjects: Q Science > Q Science (General) > Q334-342 Computer science. Artificial intelligence. Algorithms. Robotics. Automation.
Divisions: 09-Faculty of Computer Science > 56201-Computer Systems (S1)
Depositing User: Dr. Sukemi Sukemi
Date Deposited: 18 Jan 2022 05:59
Last Modified: 18 Jan 2022 05:59
URI: http://repository.unsri.ac.id/id/eprint/60738

Actions (login required)

View Item View Item