• Journal of Korea Water Resources Association
• /
• v.57 no.3
• /
• pp.181-193
• /
• 2024

The increasing atmospheric imbalance caused by climate change leads to an elevation in precipitation, resulting in a heightened frequency of flooding. Consequently, there is a growing need for technology to detect and monitor these occurrences, especially as the frequency of flooding events rises. To minimize flood damage, continuous monitoring is essential, and flood areas can be detected by the Synthetic Aperture Radar (SAR) imagery, which is not affected by climate conditions. The observed data undergoes a preprocessing step, utilizing a median filter to reduce noise. Classification techniques were employed to classify water bodies and non-water bodies, with the aim of evaluating the effectiveness of each method in flood detection. In this study, the Otsu method and Support Vector Machine (SVM) technique were utilized for the classification of water bodies and non-water bodies. The overall performance of the models was assessed using a Confusion Matrix. The suitability of flood detection was evaluated by comparing the Otsu method, an optimal threshold-based classifier, with SVM, a machine learning technique that minimizes misclassifications through training. The Otsu method demonstrated suitability in delineating boundaries between water and non-water bodies but exhibited a higher rate of misclassifications due to the influence of mixed substances. Conversely, the use of SVM resulted in a lower false positive rate and proved less sensitive to mixed substances. Consequently, SVM exhibited higher accuracy under conditions excluding flooding. While the Otsu method showed slightly higher accuracy in flood conditions compared to SVM, the difference in accuracy was less than 5% (Otsu: 0.93, SVM: 0.90). However, in pre-flooding and post-flooding conditions, the accuracy difference was more than 15%, indicating that SVM is more suitable for water body and flood detection (Otsu: 0.77, SVM: 0.92). Based on the findings of this study, it is anticipated that more accurate detection of water bodies and floods could contribute to minimizing flood-related damages and losses.

• Korean Journal of Remote Sensing
• /
• v.23 no.6
• /
• pp.559-566
• /
• 2007

The generation of precise digital elevation model (DEM) is very important in coastal area where time series are especially required. Although a LIDAR system is useful in coastal regions, it is not yet popular in Korea mainly because of its high surveying cost and national security reasons. Recently, precise DEM has been made using radar interferometry and waterline methods. One of these methods, spaceborne imaging radar interferometry has been widely used to measure the topography and deformation of the Earth. We acquired ALOS PALSAR FBD mode (Fine Beam Dual) data for evaluating the quality of interferograms and their coherency. We attempted to construct DEM using ALOS PALSAR pairs - One pair is 2007/05/22 and 2007/08/22, another pair is 2007/08/22 and 2007/10/22 with respective perpendicular baseline of 820 m, 312m and respective height sensitivity of 75 m and 185m at southern of Ganghwa tidal flat, Siwha- and Hwaong-lake over west coastal of Korea peninsula. Ganghwa tidal flat has low coherence between 0.3 and 0.5 of 2007/05/22 and 2007/08/22 pair. However, Siwha-lake and Hwaong-lake areas have a higher coherence value (From 0.7 and 0.9) than Ganghwa tidal area. The reason of difference coherence value is tidal condition between tidal flat area (Ganghwa) and reclaimed zone (Siwha-lake and Hwaong-lake). Therefore, DEM was constructed by ALOS PALSAR pair over Siwha-lake and Hwaong-lake. If the temporal baseline is enough short to maintain the coherent phases and height sensitivity is enough small, we will be able to successfully construct a precise DEM over coastal area. From now on, more ALOS PALSAR data will be needed to construct precise DEM of West Coast of Korea peninsular.