Flood is one of the most disruptive natural hazards, responsible for loss of lives and damage to properties. A number of cities are subject to monsoons influences and hence face the disaster almost every year. Early notification of flood incident could benefit the authorities and public to devise both short and long terms preventive measures, to prepare evacuation and rescue mission, and to relieve the flood victims. Geographical locations of affected areas and respective severities, for instances, are among the key determinants in most flood administration. Thus far, an effective means of anticipating flood in advance remains lacking. Existing tools were typically based on manually input and prepared data. The processes were tedious and thus prohibitive for real-time and early forecasts. Furthermore, these tools did not fully exploit more comprehensive information available in current big data platforms. Therefore, this paper proposes a novel flood forecasting system based on fusing meteorological, hydrological, geospatial, and crowdsource big data in an adaptive machine learning framework. Data intelligence was driven by stateof-the-art learning strategies. Subjective and objective evaluations indicated that the developed system was able to forecast flood incidents, happening in specific areas and time frames. It was also later revealed by benchmarking experiments that the system configured with an MLP ANN gave the most effective prediction, with correct percentage, Kappa, MAE and RMSE of 97.93, 0.89, 0.01 and 0.10, respectively. INDEX TERMS Flood forecasting system, big data, machine learning, crowdsource, deep learning. PARAMATE HORKAEW received the B.Eng. degree (Hons.

No abstract

Segmenting a liver and its peripherals from abdominal computed tomography is a crucial step toward computer aided diagnosis and therapeutic intervention. Despite the recent advances in computing methods, faithfully segmenting the liver has remained a challenging task, due to indefinite boundary, intensity inhomogeneity, and anatomical variations across subjects. In this paper, a semi-automatic segmentation method based on multivariable normal distribution of liver tissues and graph-cut sub-division is presented. Although it is not fully automated, the method minimally involves human interactions. Specifically, it consists of three main stages. Firstly, a subject specific probabilistic model was built from an interior patch, surrounding a seed point specified by the user. Secondly, an iterative assignment of pixel labels was applied to gradually update the probabilistic map of the tissues based on spatio-contextual information. Finally, the graph-cut model was optimized to extract the 3D liver from the image. During post-processing, overly segmented nodal regions due to fuzzy tissue separation were removed, maintaining its correct anatomy by using robust bottleneck detection with adjacent contour constraint. The proposed system was implemented and validated on the MICCAI SLIVER07 dataset. The experimental results were benchmarked against the state-of-the-art methods, based on major clinically relevant metrics. Both visual and numerical assessments reported herein indicated that the proposed system could improve the accuracy and reliability of asymptomatic liver segmentation.

No abstract