Susanto, Divira Salsabiil (2024) Perancangan Teknologi AI Untuk Analisis dan Pemetaan Wilayah Terdampak Banjir Dengan Pendekatan Visual Komputer. Undergraduate thesis, Fakultas Teknologi Informasi Universitas Merdeka Malang.

Preview	Text HALAMAN AWAL.pdf Download (722kB) | Preview
Preview	Text BAB I.pdf Download (136kB) | Preview
	Text BAB II.pdf Restricted to Repository staff only Download (302kB)
	Text BAB III.pdf Restricted to Repository staff only Download (453kB)
	Text BAB IV.pdf Restricted to Repository staff only Download (1MB)
	Text BAB V.pdf Restricted to Repository staff only Download (31kB)
Preview	Text DAFTAR PUSTAKA.pdf Download (95kB) | Preview
Preview	Text HASIL CEK PLAGIASI.pdf Download (436kB) | Preview

Abstract

Flooding is an event that can cause an accumulation of water in areas that are not normally submerged, such as agricultural land, settlements, and city centers. Most floods are caused by the volume of water exceeding the capacity of a river or drainage system. While harmless if they do not cause loss, loss of life or injury, recurrent and significant flooding can disrupt human life and result in large economic losses. The response to flooding involves a series of steps from emergency response to future preventive measures, with both physical and non- physical approaches. Recent research has utilized big data and machine learning to reduce the impact of floods. Machine learning enables flood analysis and prediction based on historical data, which can improve the performance of prediction systems and provide cost-effective solutions. The proposed solution is to use artificial intelligence to identify areas vulnerable to flooding, facilitating the implementation of more effective preventive measures. The process includes image segmentation, validation, and model development to map areas prone to flooding. The result of this research analysis is image segmentation that can effectively identify potentially flooded areas. In flood prevention and the creation of optimal solutions as flood detection before a disaster occurs, this feature can also help disaster management agencies if they can work together in implementing more efficient and responsive emergency response strategies. This research is expected to improve emergency response efficiency, develop machine learning techniques, raise awareness about flood management, and expand knowledge about disaster management.

Actions (login required)

View Item