• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1925-1934
• /
• 2022

Monitoring nuclear activity for inaccessible areas using remote sensing technology is essential for nuclear non-proliferation. In recent years, deep learning has been actively used to detect nuclear-activity-related small objects. However, high-resolution satellite imagery containing small objects can result in class imbalance. As a result, there is a performance degradation problem in detecting small objects. Therefore, this study aims to improve detection accuracy by analyzing the effect of the ratio of small objects related to nuclear activity in the input data for the performance of the deep learning model. To this end, six case datasets with different ratios of small object pixels were generated and a U-Net model was trained for each case. Following that, each trained model was evaluated quantitatively and qualitatively using a test dataset containing various types of small object classes. The results of this study confirm that when the ratio of object pixels in the input image is adjusted, small objects related to nuclear activity can be detected efficiently. This study suggests that the performance of deep learning can be improved by adjusting the object pixel ratio of input data in the training dataset.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_1
• /
• pp.597-608
• /
• 2022

Remotely sensed data have been used in various fields, such as disasters, agriculture, urban planning, and the military. Recently, the demand for the multitemporal dataset with the high-spatial-resolution has increased. This manuscript proposed an automatic image matching algorithm using a deep learning technique to utilize a multitemporal remotely sensed dataset. The proposed deep learning model was based on High Resolution Net (HRNet), widely used in image segmentation. In this manuscript, denseblock was added to calculate the correlation map between images effectively and to increase learning efficiency. The training of the proposed model was performed using the multitemporal orthophotos of the National Geographic Information Institute (NGII). In order to evaluate the performance of image matching using a deep learning model, a comparative evaluation was performed. As a result of the experiment, the average horizontal error of the proposed algorithm based on 80% of the image matching rate was 3 pixels. At the same time, that of the Zero Normalized Cross-Correlation (ZNCC) was 25 pixels. In particular, it was confirmed that the proposed method is effective even in mountainous and farmland areas where the image changes according to vegetation growth. Therefore, it is expected that the proposed deep learning algorithm can perform relative image registration and image matching of a multitemporal remote sensed dataset.

• Journal of Korea Water Resources Association
• /
• v.54 no.7
• /
• pp.453-462
• /
• 2021

As the frequency of localized heavy rainfall has increased during recent years, the importance of high-resolution radar data has also increased. This study aims to correct the bias of Dual Polarization radar that still has a spatial and temporal bias. In many studies, various statistical techniques have been attempted to correct the bias of radar rainfall. In this study, the bias correction of the S-band Dual Polarization radar used in flood forecasting of ME was implemented by a Convolutional Autoencoder (CAE) algorithm, which is a type of Convolutional Neural Network (CNN). The CAE model was trained based on radar data sets that have a 10-min temporal resolution for the July 2017 flood event in Cheongju. The results showed that the newly developed CAE model provided improved simulation results in time and space by reducing the bias of raw radar rainfall. Therefore, the CAE model, which learns the spatial relationship between each adjacent grid, can be used for real-time updates of grid-based climate data generated by radar and satellites.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.4
• /
• pp.359-364
• /
• 2022

The change detection in remote sensing results in errors due to the camera's optical factors, seasonal factors, and land cover characteristics. The inclination of the building in the image was simulated according to the camera angle using the Cycle Generative Adversarial Network method, and the simulated image was used to contribute to the improvement of change detection accuracy. Based on CycleGAN, the inclination of the building was similarly simulated to the building in the other image based on the image of one of the two periods, and the error of the original image and the inclination of the building was compared and analyzed. The experimental data were taken at different times at different angles, and Kompsat-3A high-resolution satellite images including urban areas with dense buildings were used. As a result of the experiment, the number of incorrect detection pixels per building in the two images for the building area in the image was shown to be reduced by approximately 7 times from 12,632 in the original image and 1,730 in the CycleGAN-based simulation image. Therefore, it was confirmed that the proposed method can reduce detection errors due to the inclination of the building.

• Korean Journal of Remote Sensing
• /
• v.39 no.2
• /
• pp.157-168
• /
• 2023

Image segmentation and supervised classification techniques are widely used to monitor the ground surface using high-resolution remote sensing images. In order to classify various objects, a process of defining a class corresponding to each object and selecting samples belonging to each class is required. Existing methods for extracting class samples should select a sufficient number of samples having similar intensity characteristics for each class. This process depends on the user's visual identification and takes a lot of time. Representative samples of the class extracted are likely to vary depending on the user, and as a result, the classification performance is greatly affected by the class sample extraction result. In this study, we propose an image classification technique that minimizes user intervention when extracting class samples by applying the histogram back-projection technique and has consistent intensity characteristics of samples belonging to classes. The proposed classification technique using histogram back-projection showed improved classification accuracy in both the experiment using hue subchannels of the hue saturation value transformed image from Compact Advanced Satellite 500-1 imagery and the experiment using the original image compared to the technique that did not use histogram back-projection.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.34 no.1
• /
• pp.43-52
• /
• 2016

Due to the characteristics of microwave signals, Radar satellite image has been used for flood detection without weather and time influence. The more methods of flood detection were developed, the more detection rate of flood area has been increased. Since flood causes a lot of damages, flooded area should be distinguished from non flooded area. Also, the detection of flood area should be accurate. Therefore, not only image resolution but also the filtering process is critical to minimize resolution degradation. Although a resolution of radar images become better as technology develops, there were a limited focused on a highly suitable filtering methods for flood detection. Thus, the purpose of this study is to find out the most appropriate filtering method for flood detection by comparing three filtering methods: Lee filter, Frost filter and NL-means filter. Therefore, to compare the filters to detect floods, each filters are applied to the radar image. Comparison was drawn among filtered images. Then, the flood map, results of filtered images are compared in that order. As a result, Frost and NL-means filter are more effective in removing the speckle noise compared to Lee filter. In case of Frost filter, resolution degradation occurred severly during removal of the noise. In case of NL-means filter, shadow effect which could be one of the main reasons that causes false detection were not eliminated comparing to other filters. Nevertheless, result of NL-means filter shows the best detection rate because the number of shadow pixels is relatively low in entire image. Kappa coefficient is scored 0.81 for NL-means filtered image and 0.55, 0.64 and 0.74 follows for non filtered image, Lee filtered image and Frost filtered image respectively. Also, in the process of NL-means filter, speckle noise could be removed without resolution degradation. Accordingly, flooded area could be distinguished effectively from other area in NL-means filtered image.

• Korean Journal of Remote Sensing
• /
• v.36 no.6_2
• /
• pp.1509-1521
• /
• 2020

Experiments for validation of surface reflectance produced by Korea Multi-Purpose Satellite (KOMPSAT-3A) were conducted using Chinese Baotou (BTCN) data among four sites of the Radical Calibration Network (RadCalNet), a portal that provides spectrophotometric reflectance measurements. The atmosphere reflectance and surface reflectance products were generated using an extension program of an open-source Orfeo ToolBox (OTB), which was redesigned and implemented to extract those reflectance products in batches. Three image data sets of 2016, 2017, and 2018 were taken into account of the two sensor model variability, ver. 1.4 released in 2017 and ver. 1.5 in 2019, such as gain and offset applied to the absolute atmospheric correction. The results of applying these sensor model variables showed that the reflectance products by ver. 1.4 were relatively well-matched with RadCalNet BTCN data, compared to ones by ver. 1.5. On the other hand, the reflectance products obtained from the Landsat-8 by the USGS LaSRC algorithm and Sentinel-2B images using the SNAP Sen2Cor program were used to quantitatively verify the differences in those of KOMPSAT-3A. Based on the RadCalNet BTCN data, the differences between the surface reflectance of KOMPSAT-3A image were shown to be highly consistent with B band as -0.031 to 0.034, G band as -0.001 to 0.055, R band as -0.072 to 0.037, and NIR band as -0.060 to 0.022. The surface reflectance of KOMPSAT-3A also indicated the accuracy level for further applications, compared to those of Landsat-8 and Sentinel-2B images. The results of this study are meaningful in confirming the applicability of Analysis Ready Data (ARD) to the surface reflectance on high-resolution satellites.

• Korean Journal of Remote Sensing
• /
• v.14 no.3
• /
• pp.296-306
• /
• 1998

Airborne MSS data is regarded as a potentially effective data source for the measurement of water quality and for the environmental change of water bodies. In this study, we measured the radiance reflectance by using multi-spectral image of low resolution camera(LRC) which will be reached in the multi-purpose satellite(KOMPSAT) to use the data in analyzing water pollution. We also investigated the possibility of extraction of water quality factors in water bodies by using high resolution remote sensing data such as Airborne MSS. Especially, we tried to extract environmental factors related with eutrophication such as chlorophyll-a, suspended sediments and turbidity, and also tried to develop the process technique and the radiance feature of reflectance related with eutrophication. Although it was difficult to explicitly correlate Airborne MSS data with water quality factors due to the insufficient number of ground truth data. The results were summarized as follows: First, the spectrum of sun's rays which reaches the surface of the earth was consistent with visible bands of 0.4${\mu}{\textrm}{m}$~0.7${\mu}{\textrm}{m}$ and about 50% of total quantity of radiation could be found. The spectrum was reached highest at around 0.5${\mu}{\textrm}{m}$ of green spectral band in visible bands. Second, as a result of the radiance reflectance Chlorophyll-a represented high mainly around 0.52${\mu}{\textrm}{m}$ of green spectral band, and suspended sediments and turbidity represented high at 0.8${\mu}{\textrm}{m}$ and at 0.57${\mu}{\textrm}{m}$, respectively. Finally, as a result of the water quality analysis by using Airborne MSS, Chlorophyll-a could have a distribution image after carrying out ratio of B3 and B5 to B7. Band 7 was useful for making the distribution image of suspended sediments. When we carried out PCA, suspended sediments and turbidity had distributions at PC 1 and PC 4 which are similar to the ground data. Above results can be changed according to the change of season and time. Therefore, in order to analyze the environmental factors of water quality by using LRC data more exactly, we need to investigate the ground data and the radiance feature of reflectance of water bodies constantly. For further studies, we will constantly analyze the radiance feature of the surface of water in wafter bodies by measuring the on-the-spot radiance reflectance and using low resolution satellite image(SeaWiFS). We will also gather the data of water quality analysis in water bodies and analyze the pattern of water pollution.

• Journal of the Korean Association of Geographic Information Studies
• /
• v.21 no.1
• /
• pp.35-45
• /
• 2018

Image data collected by an airborne hyperspectral camera system have a great usability in coastal line mapping, detection of facilities composed of specific materials, detailed land use analysis, change monitoring and so forh in a complex coastal area because the system provides almost complete spectral and spatial information for each image pixel of tens to hundreds of spectral bands. A few approaches after classifying by a few approaches based on SAM(Spectral Angle Mapper) supervised classification were applied for extracting optimal land cover information from hyperspectral images acquired by CASI-1500 airborne hyperspectral camera on the object of a coastal area which includes both land and sea water areas. We applied three different approaches, that is to say firstly the classification approach of combined land and sea areas, secondly the reclassification approach after decompostion of land and sea areas from classification result of combined land and sea areas, and thirdly the land area-only classification approach using atmospheric correction images and compared classification results and accuracies. Land cover classification was conducted respectively by selecting not only four band images with the same wavelength range as IKONOS, QuickBird, KOMPSAT and GeoEye satelllite images but also eight band images with the same wavelength range as WorldView-2 from 48 band hyperspectral images and then compared with the classification result conducted with all of 48 band images. As a result, the reclassification approach after decompostion of land and sea areas from classification result of combined land and sea areas is more effective than classification approach of combined land and sea areas. It is showed the bigger the number of bands, the higher accuracy and reliability in the reclassification approach referred above. The results of higher spectral resolution showed asphalt or concrete roads was able to be classified more accurately.

• Korean Journal of Remote Sensing
• /
• v.37 no.5_1
• /
• pp.1177-1186
• /
• 2021

As the Arctic melt ponds play an important role in determining the interannual variation of the sea ice extent and changes in the Arctic environment, it is crucial to monitor the Arctic melt ponds with high accuracy. Ice, Cloud, and Land Elevation Satellite-2 (ICESat-2), which is the NASA's latest altimeter satellite based on the green laser (532 nm), observes the global surface elevation. When compared to the CryoSat-2 altimetry satellite whose along-track resolution is 250 m, ICESat-2 is highly expected to provide much more detailed information about Arctic melt ponds thanks to its high along-track resolution of 70 cm. The basic products of ICESat-2 are the surface height and the number of reflected photons. To aggregate the neighboring information of a specific ICESat-2 photon, the segments of photons with 10 m length were used. The standard deviation of the height and the total number of photons were calculated for each segment. As the melt ponds have the smoother surface than the sea ice, the lower variation of the height over melt ponds can make the melt ponds distinguished from the sea ice. When the melt ponds were extracted, the number of photons per segment was used to classify the melt ponds covered with open-water and specular ice. As photons are much more absorbed in the water-covered melt pondsthan the melt ponds with the specular ice, the number of photons persegment can distinguish the water- and ice-covered ponds. As a result, the suggested melt pond detection method was able to classify the sea ice, water-covered melt ponds, and ice-covered melt ponds. A qualitative analysis was conducted using the Sentinel-2 optical imagery. The suggested method successfully classified the water- and ice-covered ponds which were difficult to distinguish with Sentinel-2 optical images. Lastly, the pros and cons of the melt pond detection using satellite altimetry and optical images were discussed.