Research Article
Ensemble Classification Technique for Cultural Heritage Image
@INPROCEEDINGS{10.1007/978-3-031-04409-0_3, author={Sathit Prasomphan}, title={Ensemble Classification Technique for Cultural Heritage Image}, proceedings={Machine Learning and Intelligent Communications. 6th EAI International Conference, MLICOM 2021, Virtual Event, November 2021, Proceedings}, proceedings_a={MLICOM}, year={2022}, month={5}, keywords={Cultural heritage Content retrieval Ensemble classification technique Machine learning Image processing}, doi={10.1007/978-3-031-04409-0_3} }- Sathit Prasomphan
Year: 2022
Ensemble Classification Technique for Cultural Heritage Image
MLICOM
Springer
DOI: 10.1007/978-3-031-04409-0_3
Abstract
This research aims to propose an ensemble classification technique for cultural heritage image. In the case of Thai architecture, the cultural heritage content was extracted. The key contribution of this study was the creation of an ensemble classification technique for recovering information from a cultural heritage images in order to convey a story inside it. Through the learning of machine learning and image processing, the cultural heritage of Thai archaeological site architecture was generated, including the story of the archaeological site. The significant information contained inside the cultural heritage image was extracted and presented to those who are interested in its contents. The appearance of the form within an image can be used to differentiate image qualities such as era, architecture, and style. The classification result of the ensemble classification technique is used to examine the experimental findings of cultural heritage content retrieval. Photographs of Thai archaeological site architecture from world heritage provinces in Thailand were categorized in this investigation, for example, images from Phra Nakhon Si Ayutta province, Sukhothai province, and Bangkok, which represent the Ayutthaya, Sukhothai, and Rattanakosin eras in sequential sequence. On average, the proposed ensemble classification technique does have an accuracy of 80.83%.
