Iswari, Ni Made Satvika and Wella, Wella and Ranny, Ranny (2017) Fish freshness classification method based on fish image using k-Nearest Neighbor. 2017 4th International Conference on New Media Studies (CONMEDIA).
Full text not available from this repository.Abstract
The potential of fish production in Indonesia is very high because of the Indonesia territory which consists of waters (sea, lake, river, and pond). However, fish consumption in the community is still very low. Public awareness of the freshness of fish consumed also become an existing problem. Communities need facilities that can be used easily and accurately in choosing a fish worth consumption, because not infrequently the process of decomposition of fish is not realized by the fish distributors. In addition, the lack of the use of technology makes fishing production run slowly. The process of sorting fish manually makes the fish freshness that reaches consumer hands cannot be ascertained. In this research, a method to classify the fish freshness based on fish image was developed. k-Nearest Neighbour (kNN) was used as the classification algorithm based on fish image colours summarization. Accuracy result of the classification by using kNN was 91.36%. This indicates that the resulting method was acceptable. Meanwhile, the colour that determines the fish freshness the most was the black colour of the fish eyes. It was because the black colour had the highest Information Gain for all type of the fish used.
| Item Type: | Article |
|---|---|
| Subjects: | 000 Computer Science, Information and General Works > 000 Computer Science, Knowledge and Systems > 006 Special Computer Methods 300 Social Sciences > 330 Economics > 333 Economics of Land and Energy |
| Divisions: | Faculty of Engineering & Informatics > Informatics |
| Depositing User: | Administrator UMN Library |
| Date Deposited: | 11 Oct 2021 09:04 |
| Last Modified: | 11 Oct 2021 09:04 |
| URI: | https://kc.umn.ac.id/id/eprint/18625 |
Actions (login required)
 |
View Item |
