Evidence record · UDINUS · SDGs
- Unit
- Rektorat UDINUS
- Period
- 2020
- Goals
- Record
- fbb0bbec-c8ef-4a18-bfa3-8e7ab0e43e3e
- Published
- 2026-10-30
Universitas Dian Nuswantoro (UDINUS) conducted a study on the automatic classification of Ribbed Smoked Sheets (RSS) based on the presence or absence of mold. The activity participants were Arifin, Zaenal; Yuni Madya Mukti, Eki; Jehan Tamamy, Aries; and Ary Heryanto, M. The study concerned natural rubber processing in the form of smoked rubber sheets, known as Ribbed Smoked Sheets. Indonesia is an exporter of natural rubber, and RSS is one type of processed rubber used as export material. The source states that RSS quality greatly affects the increase in rubber exports.
The quality of RSS has been stipulated in SNI 06-001-1987 and the International Standards of Quality and Packing for Natural Rubber Grades, also known as The Green Book. The process of determining RSS quality is called sorting. In some rubber plantations, sorting is still performed manually by observing the presence or absence of mold on the surface of RSS with the naked eye. According to the source, this manual process can produce inaccurate and subjective quality results.
To address this sorting process, the study carried out an automatic classification based on the presence of mold on RSS. The research used image processing, with a rubber sheet image as the input and classification results as the output. The classification used the Neural Network Perceptron method and produced two categories: moldy RSS and non-moldy RSS. The study used 1000 RSS images as training data and 50 smoked rubber sheet images as test data. The source also states that each test had 100 pieces, with an accuracy value of 96% and the best epoch value on the 4th epoch.
In relation to SDG 17, the activity represents SDG-related research involving image processing and the Neural Network Perceptron method for RSS classification. The available source supports the research topic, its sorting-related problem, the stated method, the two classification categories, and the reported results. No further information about institutional partnerships or other SDG 17 activities is provided in the source.