Geo Informatics and Space Technology Development Agency Public Organization

Bangkok, Thailand

Geo Informatics and Space Technology Development Agency Public Organization

Bangkok, Thailand
SEARCH FILTERS
Time filter
Source Type

Suwanlertcharoen T.,Geo Informatics and Space Technology Development Agency Public Organization | Pongput K.,Kasetsart University
International Journal of Geoinformatics | Year: 2017

This study aims to examine the riverbank changes and the impacts of water level rise on the right bank of the Mekong River in Nong Khai province, northeast Thailand. The Normalized Difference Water Index (NDWI) technique was applied for the extraction of water features from LANDSAT imagery to detect the change. The result shows both erosion and accretion in the study area during the dry season between the year 2000 and 2014 analyzed by LANDSAT-7 TM images recorded on February 13, 2000 and LANDSAT-8 OLI images recorded on March 31, 2014. The total area of erosion is 0.89 sq.km or 555.90 Rais at average rate of 1.31 m/year whereas total area of accretion is 3.93 sq.km or 2,457.27 Rais at an average rate of2 03 m/year. The morphological change of sandbars and river islets finds eroded/submerged area of 5.92 sq.km or 3,698.58 Rais and accreted area of 1.85 sq. bn or 1,153.21 Rais. Digital Elevation Model (DEM) is applied together with the Mekong River channel data year 2014 to simulate the water level rise scenarios. The result identifies that agricultural land would be the most at risk of flooding, next are miscellaneous land (water body, sandbar, islet, etc.) and forest land respectively. © Geoinformatics International.


Silva K.,Thailand Institute of Nuclear Technology | Lawawirojwong S.,Geo Informatics and Space Technology Development Agency Public Organization | Promping J.,Thailand Institute of Nuclear Technology
Journal of Physics: Conference Series | Year: 2017

Consequence assessment of a hypothetical severe accident is one of the important elements of the risk assessment of a nuclear power plant. It is widely known that the meteorological conditions can significantly influence the outcomes of such assessment, since it determines the results of the calculation of the radionuclide environmental transport. This study aims to assess the impacts of the meteorological conditions to the results of the consequence assessment. The consequence assessment code, OSCAAR, of Japan Atomic Energy Agency (JAEA) is used for the assessment. The results of the consequence assessment using Thai meteorological data are compared with those using Japanese meteorological data. The Thai case has following characteristics. Low wind speed made the radionuclides concentrate at the center comparing to the Japanese case. The squalls induced the peaks in the ground concentration distribution. The evacuated land is larger than the Japanese case though the relocated land is smaller, which is attributed to the concentration of the radionuclides near the release point. © Published under licence by IOP Publishing Ltd.


Panboonyuen T.,Chulalongkorn University | Vateekul P.,Chulalongkorn University | Jitkajornwanich K.,King Mongkut's University of Technology Thonburi | Lawawirojwong S.,Geo Informatics and Space Technology Development Agency Public Organization
Advances in Intelligent Systems and Computing | Year: 2018

Object classification from images is among the many practical examples where deep learning algorithms have successfully been applied. In this paper, we present an improved deep convolutional encoder-decoder network (DCED) for segmenting road objects from aerial images. Several aspects of the proposed method are enhanced, incl. incorporation of ELU (exponential linear unit)—as opposed to ReLU (rectified linear unit) that typically outperforms ELU in most object classification cases; amplification of datasets by adding incrementally-rotated images with eight different angles in the training corpus (this eliminates the limitation that the number of training aerial images is usually limited), thus the number of training datasets is increased by eight times; and lastly, adoption of landscape metrics to further improve the overall quality of results by removing false road objects. The most recent DCED approach for object segmentation, namely SegNet, is used as one of the benchmarks in evaluating our method. The experiments were conducted on a well-known aerial imagery, Massachusetts roads dataset (Mass. Roads), which is publicly available. The results showed that our method outperforms all of the baselines in terms of precision, recall, and F1 scores. © Springer International Publishing AG 2018.


Sritarapipat T.,Geo Informatics and Space Technology Development Agency Public Organization | Rakwatin P.,Geo Informatics and Space Technology Development Agency Public Organization | Kasetkasem T.,Kasetsart University
Sensors (Basel, Switzerland) | Year: 2014

Rice crop height is an important agronomic trait linked to plant type and yield potential. This research developed an automatic image processing technique to detect rice crop height based on images taken by a digital camera attached to a field server. The camera acquires rice paddy images daily at a consistent time of day. The images include the rice plants and a marker bar used to provide a height reference. The rice crop height can be indirectly measured from the images by measuring the height of the marker bar compared to the height of the initial marker bar. Four digital image processing steps are employed to automatically measure the rice crop height: band selection, filtering, thresholding, and height measurement. Band selection is used to remove redundant features. Filtering extracts significant features of the marker bar. The thresholding method is applied to separate objects and boundaries of the marker bar versus other areas. The marker bar is detected and compared with the initial marker bar to measure the rice crop height. Our experiment used a field server with a digital camera to continuously monitor a rice field located in Suphanburi Province, Thailand. The experimental results show that the proposed method measures rice crop height effectively, with no human intervention required.


Chaimatanan S.,Geo Informatics and Space Technology Development Agency Public Organization | Vongsantivanich W.,Geo Informatics and Space Technology Development Agency Public Organization
ACRS 2015 - 36th Asian Conference on Remote Sensing: Fostering Resilient Growth in Asia, Proceedings | Year: 2015

Mission planning plays a crucial role for space-based remote sensing, as it is one of the key processes that defines the success rate of each daily mission. From user's request, the Earth observation satellite mission-planner concerns with scheduling the acquisition plan that will be used to prepare the satellite command queue to perform the image acquisitions. On a long-term basis, the mission-plan should optimize a set of pre-defined goals, while on a daily basis the mission plan must be adapted to the observation condition. This paper presents a mission-planning tool that is being developed for Thaichote Earth observation satellite to improve and optimize Gistda's space-based remote sensing resources. It is designed to take into account the problems from a real operation point of view, where rescheduling is necessary due to the change of observation conditions or the requirements for acquisition modification, while the changes to the long-term goal scheduling should be minimized. The tool allows the mission-planner to manage the utilization of the satellite in four different time levels; long-term, medium-term, short-term, and post-mission. It consists of four main modules, which are user interface, feasibility assessment, planning optimization, and data management modules. It was implemented and tested with real input data based on Thaichote operation. The perspective to use this tool for satellite constellation and regional area acquisition is also discussed at the end of this paper.


Tangpattanakul P.,Geo Informatics and Space Technology Development Agency Public Organization
2015 4th International Conference on Informatics, Electronics and Vision, ICIEV 2015 | Year: 2015

This paper presents two algorithms, which are a nondominated sorting genetic algorithm II (NSGA-II) and an indicator-based multi-objective local search (IBMOLS), for solving a bi-objective p-Median problem. The bi-objective p-Median problem is a problem of finding p location points to install facilities from a set of m candidates. This problem considers two objectives: minimizing the sum of the distances from each customer to the nearest facility and minimizing the sum of the costs to install each facility in the selected location points. NSGA-II and IBMOLS are efficient algorithms in the area of multi-objective optimization. Experiments are conducted on generated instances. Hypervolume values of the approximate Pareto fronts are computed and the obtained results from IBMOLS and NSGA-II are compared. © 2015 IEEE.


Srestasathiern P.,Geo Informatics and Space Technology Development Agency Public Organization | Rakwatin P.,Geo Informatics and Space Technology Development Agency Public Organization
Remote Sensing | Year: 2014

Oil palm tree is an important cash crop in Thailand. To maximize the productivity from planting, oil palm plantation managers need to know the number of oil palm trees in the plantation area. In order to obtain this information, an approach for palm tree detection using high resolution satellite images is proposed. This approach makes it possible to count the number of oil palm trees in a plantation. The process begins with the selection of the vegetation index having the highest discriminating power between oil palm trees and background. The index having highest discriminating power is then used as the primary feature for palm tree detection. We hypothesize that oil palm trees are located at the local peak within the oil palm area. To enhance the separability between oil palm tree crowns and background, the rank transformation is applied to the index image. The local peak on the enhanced index image is then detected by using the non-maximal suppression algorithm. Since both rank transformation and non-maximal suppression are window based, semi-variogram analysis is used to determine the appropriate window size. The performance of the proposed method was tested on high resolution satellite images. In general, our approach uses produced very accurate results, e.g., about 90 percent detection rate when compared with manual labeling. © 2014 by the authors.


Tangpattanakul P.,Geo Informatics and Space Technology Development Agency Public Organization | Chaimatanan S.,Geo Informatics and Space Technology Development Agency Public Organization
ACRS 2015 - 36th Asian Conference on Remote Sensing: Fostering Resilient Growth in Asia, Proceedings | Year: 2015

This work proposes a biased random key genetic algorithm (BRKGA) for solving a multiple priorities task scheduling problem. Generally, some chromosomes are initially generated by encoding as key vectors of real value between [0, 1] in BRKGA process. To obtain the solution, the chromosome has to be decoded. The multiple priorities task scheduling problem, which needs to schedule strips of Earth observation before sending the sequence to the satellite, is considered in this work. Strips are required from customers with different priorities and each strip has its own priority. The objective of this scheduling problem is to minimize the finishing date for taking all strips. For encoding, each gene in the chromosome represents each strip. The gene value of each strip is a real number in different intervals depending on the priority of related strip. Let si be a strip with priority P[i], where i ϵ {1, 2,⋯, n} and P[i] € {1, 2,⋯, np}, n is number of strips and np is number of different priorities. Thus, gene values of strip si with priority P[i] is a real value in the interval [(P[i]-1)/np, P[i]/np). For decoding, the sequence of strip acquisition can be obtained from the represented gene values. The strip with the lowest gene value is considered firstly. It is assigned in the sequence of strip acquisition as early as possible and it must also satisfy the constraints. Then, the strip with the next higher gene value is considered and it can be inserted to the available spaces of the sequence that satisfy the constraints. Moreover, the considered strip is assigned in the sequence as early as possible. Experiments are conducted on a realistic instance, which concerns strips with two different priorities. All strips are assigned in the sequence and then, this sequence will be sent to the Earth observing satellite.


Jitkajornwanich K.,Geo Informatics and Space Technology Development Agency Public Organization | Elmasri R.,University of Texas at Arlington
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2015

Large amounts of "big data" are generated every day, many in a "raw" format that is difficult to analyze and mine. This data contains potential hidden meaningful concepts, but much of the data is superfluous and not of interest to the domain experts. Thus, dealing with big raw data solely by applying a set of distributed computing technologies (e.g., MapReduce, BSP [Bulk Synchronous Parallel], and Spark) and/or distributed storage systems, namely NoSQL, is generally not sufficient. Extracting the full knowledge that is hidden in the raw data is necessary to efficiently enable analysis and mining. The data needs to be processed to remove the superfluous parts and generate the meaningful domainspecific concepts. In this paper, we propose a framework that incorporates conceptual modeling and EER principle to effectively extract conceptual knowl‐ edge from the raw data so that mining and analysis can be applied to the extracted conceptual data. © Springer International Publishing Switzerland 2015.


Tawala J.,Geo Informatics and Space Technology Development Agency Public Organization
34th Asian Conference on Remote Sensing 2013, ACRS 2013 | Year: 2013

Car robbery is one the serious problems in a large city like Bangkok, the capital of Thailand, and statistics showed that one or two vehicles were stolen every day. The objective of this study is to identify risk area for car robbery in Lat Phrao district of Bangkok, The excel file of 5-year period car stolen records collected during 2005 - 2009 by Lat Phrao police station were converted into GIS format, to identify and describe spatial and temporal distribution of the incident. Next, the data were spatially interpolated using various algorithms, and the outputs were compared to select the most appropriate output. The result showed that the kernel density estimation described spatial distribution of the robbery very well, and Bang Kapi and Happy Land sub districts had the highest risk for car robbery. To make the output more understandable for general users, the kernel density estimation was overlaid on high resolution satellite imageries in Google Earth. The results provide a basis for further study on risk areas for the car robbery, analysis of factors leading to the risk, and help law enforcement administrators in making better decisions.

Loading Geo Informatics and Space Technology Development Agency Public Organization collaborators
Loading Geo Informatics and Space Technology Development Agency Public Organization collaborators