Automated Image Identification Method for Flood Disaster Monitoring In Riverine Environments: a Case Study in Taiwan

Abstract

As the globe undergoes extreme climate changes, disaster events and their disaster scale continue to increase; therefore, it has become imperative to devise disaster prevention measures. Traditional flood monitoring devices, while operating in harsh environments, are often influenced by changes in weather conditions such light, rain, and fog. Consequently, recorded images are often blurred or damaged, which increases the possibility of errors in judgment or delays in the hazard mitigation process. In this study, an automated identification method for flood monitoring based on real-time video images is proposed. The method can be used by the Water Conservancy Agency of the Ministry of Economic Affairs automated for automated identification of flooded areas and for automated determination of the water levels of the main rivers from image data. This paper presents results to show that the capability to detect temporal changes in image sequences is crucial for an automated image-based flood alarm system used in disaster-monitoring applications.

A performance analysis for real-time flood monitoring using image-based processing

“…Then define w as the adjacency matrix of the graph and let be the diagonal matrix with diagonal entries in formula (4). The cost of normalized cuts can be written in (5). is the indicator vector, where if the node is in set and otherwise, .…”

“…It"s vital to monitor the state of the flood for saving life and minimizing the losses. There are many methods can be used for the flood monitoring, but in the literature review from the reference [1][2][3][4][5][6][7][8][9], the image-based methods: region growing, canny edge detection and normalized cuts are better than the others based on the achievability and timeliness. These image-based methods will be made a simulation on MATLAB in the beginning and the method which has the best performance will be kept and improved.…”

“…In the reference, the related papers are [1,5] and [7]. Paper [1] proposed a method can detect the flash flood with the fast processing speed based on the stationary surveillance cameras, which can be used to the urban areas.…”

“…The process of detection includes the background subtraction, morphological closing, and canny edge detection, in which the canny edge detection has great effects on distinguishing the flood area with the others based on the edge density. Paper [5] can monitor and identify the flood automatically with the proposed method based on real-time video images. The images are captured from the digital camera.…”

Flood Forecasting GIS Water-Flow Visualization Enhancement (WaVE): A Case Study

“…RS is also increasingly being used for developing DEMs of higher resolution and greater accuracy. Today, flood inundation maps are commonly derived from passive microwave sensors or moderate resolution spectroradiometer imagery (i.e., MODIS and Landsat TM) [19] [20] [21]. Testing and experimental remote sensing systems are being set up to detect and map spatial resolution in near real-time [21], as well as monitor ongoing floods by comparing imagery with on-the-ground data information obtained from flood disasters [22].…”