Jong Geol Park

Artificial neural network ensemble-based land-cover classifiers using MODIS data

Terra and Aqua, two satellites launched by the NASA-centered International Earth Observing System project, house MODIS (moderate resolution imaging spectroradiometer) sensors. Moderate-resolution remote sensing allows the quantifying of land-surface type and extent, which can be used to monitor changes in land cover and land use for extended periods of time. In this article, we propose land-surface classification by applying an ensemble technique based on fault masking among individual classifiers in N-version programming. An N-version programming ensemble of artificial neural networks is created, in which the majority vote result is used to predict land-surface cover from MODIS data. It is shown by experiment that an N-version programming ensemble of neural networks greatly improves the classification error rate of land-cover type.

Development and evaluation of satellite image data analysis infrastructure

Tokyo University of Information Sciences (TUIS) receives moderate resolution imaging spectroradiometer (MODIS) data, and provides the processed data to universities and research institutes as part of the academic frontier project. One of the major fields of research using MODIS data is the analysis of changes in the environment. We are currently developing applications to analyze environmental changes. These applications run on our satellite image data analysis system, which is implemented in a parallel distributed system and a database server. When using satellite data, one common problem is the interference of clouds. In order to remove this interference, the standard solution is to create composite data of the same regions during a selected time span, and to patch together data which are not covered by clouds to create a clear image. We introduced a piece-processing algorithm which separates one set of satellite image data into many small pieces of image data, making it quicker and easier to analyze and process the time-series satellite data. In this research, we implemented the pieceprocessing and composite-processing algorithms in order to increase the speed of analysis within the satellite image database. We tested the proposed processing and verified its effectiveness for target applications.

Artificial neural networks paddy-field classifier using spatiotemporal remote sensing data

Monitoring changes in a paddy-field area is important since rice is a staple food and paddy agriculture is a major cropping system in Asia. For monitoring changes in land surface, various applications using different satellites have been researched in the field of remote sensing. However, monitoring a paddy-field area with remote sensing is difficult owing to the temporal changes in the land surface, and the differences in the spatiotemporal characteristics in countries and regions. In this article, we used an artificial neural network to classify paddy-field areas using moderate resolution sensor data that includes spatiotemporal information. Our aim is to automatically generate a paddy-field classifier in order to create localized classifiers for each country and region.

Ensemble of artificial neural network based land cover classifiers using satellite data

Terra and Aqua, 2 satellites launched by the NASA-centered international Earth Observing System project, house MODIS (Moderate Resolution Imaging Spectroradiometer) sensors. Moderate resolution remote sensing allows the quantifying of land surface type and extent, which can be used to monitor changes in land cover and land use for extended periods of time. In this paper, we propose applying an ensemble technique, based on fault masking among individual classifier for N-version programming. We create an N-version programming ensemble of artificial neural networks and use the majority voting result to predict land surface cover from MODIS data. We show that an N-version programming ensemble of neural networks greatly improves the classification error rate of land cover type.

Island Activities Detected by VIIRS and Validation with AIS

A possibility to monitor the reclamation activities by remote sensing was discussed. The lights observed in the night time by Day Night Band (DNB) of Visible Infrared Imaging Radiometer Suite (VIIRS), ocean color observed in the day time by visible bands of VIIRS were the tools to monitor the surface activities, and the Automated Information System (AIS) was used to verify the types and number of vessels associated with the reclamation activities. The lights as the radiance from the surface were monitored by the object based analysis, where the object was defined as a radius of 5 km from the center of the Mischief Reef in the South China Sea (SCS). The time history of surface lights exhibited the increase of the radiance from January to May 2015 and the radiance was kept in the certain level to December 2016 with some variations. The ocean color, chlorophyll-a concentration as a proxy of sediments, showed an increase from February to June 2015 and returned to a low concentration in August 2015. According to the historical data of AIS, the number of dredgers has increased from February to August 2015 and the maximum number of dredgers was recorded in June 2015. The timing of increase of lights from surface, increase of chlorophyll-a concentration, and increase of number of vessels are consistent.

Characterization of urban heat island (UHI) changes from MODIS times series using principal component analysis (PCA): Case of Dar es Salaam City Tanzania

The urban heat island (UHI) effect describes the influence of urban surfaces on temperature patterns in urban areas as opposed to surrounding areas. Several indicators have been suggested in different studies. In this study, a procedure is presented to extract value based on polygon and characterize temporal changes using MODIS times series from two surface variables, NDVI and Land Surface Temperature (LST). The result provides empirical evidence of UHI at Dar es Salaam-City center. The difference of (2°C in Night LST) between City center and outer part area, as well as low NDVI at City Center compared to outer part was observed. 12 years trend pointed out that, all areas of city facing both LST and land use/cover changes. City center showed stronger biophysical changes in terms of NDVI (decrease in NDVI Pearson correlation −0.66 compared to outer part (decrease in NDVI Pearson correlation of −0.30).

Temporal monitoring of vessels activity using day/night band in Suomi NPP on South China Sea

In this research, we focus on vessel detection using the satellite imagery of day/night band (DNB) on Suomi NPP in order to monitor the change of vessel activity on the region of South China Sea. In this paper, we consider the relation between the temporal change of vessel activities and the events on maritime environment based on the vessel traffic density estimation using DNB. DNB is a moderate resolution (350-700m) satellite imagery but can detect the fishing light of fishery boats in night time for every day. The advantage of DNB is the continuous monitoring on wide area compared to another vessel detection and locating system. However, DNB gave strong influence of cloud and lunar refection. Therefore, we additionally used Brightness Temperature at 3.7μm(BT3.7) for cloud information. In our previous research, we construct an empirical vessel detection model that based on the DNB contrast and the estimation of cloud condition using BT3.7. Moreover, we proposed a vessel traffic density estimation method based on empirical model. In this paper, we construct the time temporal density estimation map on South China Sea and East China Sea in order to extract the knowledge from vessel activities change.

Application of neural network swarm optimization for paddy-field classification from remote sensing data

Monitoring changes in paddy areas is important for economic and environmental research, since rice is a staple food in Asia and paddy agriculture is a major cropping system. Recently, remote sensing has been used to observe changes in the areas of paddy. However, monitoring paddy areas by remote sensing is difficult owing to the temporal changes in paddy, and the differences in the spatiotemporal characteristics of paddy agriculture between countries or regions. In our previous research using a multilayered perceptron and spatiotemporal satellite sensor data, the proposed classifier yielded a correct classification rate of 90.8%. In this article, we proposed a cooperative learning method using particle swarm optimization as the global search method and a multilayered perceptron as the local search method in order to improve the classification accuracy for practical use.

Applying fuzzy sets to composite algorithm for remote sensing data

Remote sensing of the earth surface using satellite mounted sensor data is a major method for global environmental monitoring today. However, when using satellite sensor data, clouds in the atmosphere can interfere with the readings, and specific land points may not be correctly monitored on a given day. In order to overcome this problem, multiple day composite data is frequently used. Multiple day composite data uses several consecutive days remote sensing data, and picks the most accurate data within the temporal dataset for the same land point. This allows creating a more complete dataset by patching together data not interfered by clouds during a specified time period, to create a clearer, more usable dataset. In this paper, we propose applying fuzzy set logic in order to select the clearest data in the temporal interval for the composite data. Moderate resolution remote sensing data of areas in Japan were used for evaluation.

Applying soft computing for remote sensing data composite algorithms

Remote sensing of the earth’s surface using satellite-mounted sensor data is one of the most important methods for global environmental monitoring today. However, when using satellite sensor data, clouds in the atmosphere can interfere with the remote sensing, and specific land points may not be correctly monitored on any given day. In order to overcome this problem, a common alternative is to use multiple day composite data. Multiple day composite data use several consecutive days’ remote sensing data, and choose the most accurate data within the temporal dataset for the same land point. This allows the creation of a more complete dataset by patching together data which have had no cloud interference during a specified time period in order to create a clearer, more usable dataset. In this article, we propose the application of soft computing, namely fuzzy logic, in order to select the clearest data in the temporal interval to use for the composite data. Moderate resolution remote sensing data of areas in Japan were used for the evaluation, and the results were compared with previous composite methods.