Three-Dimensional Convolutional Neural Network on Multi-Temporal Synthetic Aperture Radar Images for Urban Flood Potential Mapping in Jakarta

Abstract

Flooding in urban areas is counted as a significant disaster that must be correctly mitigated due to the huge amount of affected people, material losses, hampered economic activity, and flood-related diseases. One of the technologies available for disaster mitigation and prevention is satellites providing image data on previously flooded areas. In most cases, floods occur in conjunction with heavy rain. Thus, from a satellite’s optical sensor, the flood area is mostly covered with clouds which indicates ineffective observation. One solution to this problem is to use Synthetic Aperture Radar (SAR) sensors by observing backscatter differences before and after flood events. This research proposes mapping the flood-prone areas using machine learning to classify the areas using the 3D CNN method. The method was applied on a combination of co-polarized and cross-polarized SAR multi-temporal image datasets covering Jakarta City and the coastal area of Bekasi Regency. Testing with multiple combinations of training/testing data proportion split and a different number of epochs gave the optimum performance at an 80/20 split with 150 epochs achieving an overall accuracy of 0.71 after training in 283 min.

Pemetaan Daerah Rawan Banjir di Kota Ambon Menggunakan Sistim Informasi Geografis

“…Perubahan iklim yang terjadi membuat meningkatnya frekuensi kejadian bencana alam di Indonesia diantaranya banjir (Klipper et al, 2021). Selain bencana lonsor banjir adalah bencana alam yang merugikan, terutama di daerah perkotaan seperti Ambon, karena berdampak pada sejumlah besar penduduk seperti kerugian material akibat kerusakan harta benda akibat genangan banjir (Riyanto et al, 2022) . Banjir adalah bencana alam yang terjadi hampir setiap tahun terutama pada wilayah perkotaan (Amira et al, 2020).…”

Evaluation of land resilience against natural disasters using ecosystem services approach in Kendari City, Southeast Sulawesi, Indonesia

“…The results showed that areas with low resilience tended to have high multi-disaster hazards. Figure 4 shows that inundation occurs in residential areas due to prolonged rains, while the rivers cannot accommodate the overflowing water (Riyanto, Rizkinia, Arief, & Sudiana, 2022;Shige-Eda, Akiyama, Duran, Kanaya, & Katsura, 2022). The puddle's dark brown color indicates that the upstream area's erosion process is intensive, so various forms of prevention, such as reforestation or the construction and checking dams, need to be carried out as soon as possible (Dunn & Minderhoud, 2022;Singh et al, 2022;Y.…”

A hybrid deep neural network with statistical learning algorithms for flood susceptibility modeling

“…The Deep Neural Network (DNN) in combination with spatial analysis can provide a strong technique for evaluating the geographical data to identify the susceptible regions more rapidly (Riyanto et al 2022). GIS has also been applied to investigate floods, and their severe damage (Wang et al 2019).…”

“…It can also be applied to integrate different targets to process the given data in complex ways (Nayak et al 2022). The coupled DNN and GIS have been widely applied in flood susceptibility assessment and modeling (Riyanto et al 2022). Various studies have confirmed that such hybridization can be applied to detect flood susceptibility regions where the technologies applying artificial intelligence in combination with spatial analysis seek to classify and process the geospatial data in a way that goes beyond simple input/output protocols.…”