• Korean Journal of Remote Sensing
• /
• v.38 no.5_1
• /
• pp.535-543
• /
• 2022

A drone is used to diagnose crop growth and to provide information through images in the agriculture field. In the case of using high spatial resolution drone images, growth information for each object can be produced. However, accurate object detection is required and adjacent objects should be efficiently classified. The purpose of this study is to develop a Chinese cabbage object detection algorithm using multispectral reflectance images observed from drone and computer vision techniques. Drone images were captured between 7 and 15 days after planting a Chinese cabbage from 2018 to 2020 years. The thresholds of object detection algorithm were set based on 2019 year, and the algorithm was evaluated based on images in 2018 and 2019 years. The vegetation area was classified using the characteristics of spectral reflectance. Then, morphology techniques such as dilatation, erosion, and image segmentation by considering the size of the object were applied to improve the object detection accuracy in the vegetation area. The precision of the developed object detection algorithm was over 95.19%, and the recall and accuracy were over 95.4% and 93.68%, respectively. The F1-Score of the algorithm was over 0.967 for 2 years. The location information about the center of the Chinese cabbage object extracted using the developed algorithm will be used as data to provide decision-making information during the growing season of crops.

• Korean Journal of Remote Sensing
• /
• v.36 no.3
• /
• pp.401-410
• /
• 2020

Recently, many studies have been conducted on the changing phenology on the Korean Peninsula due to global warming. However, because of the geographical characteristics, research on plant season in autumn, which is difficult to measure compared to spring season, is insufficient. In this study, all leaves that maple and fallen leaves were defined as 'Decay leaves' and decay leaf detection was performed based on the Landsat-8 satellite image. The first threshold value of decay leaves was calculated by using NDVI and the secondary threshold value of decay leaves was calculated using by NDWI and the difference of spectral characteristics with green leaves. POD, FAR values were used to verify accuracy of the dry leaf detection algorithm in this study, and the results showed high accuracy with POD of 98.619 and FAR of 1.203.

• Journal of Korean Society for Geospatial Information Science
• /
• v.1 no.1 s.1
• /
• pp.193-201
• /
• 1993

In a number of areas, there are important benefits to be gained when we bring both the detection and monitoring abilities of remote sensing as well as the philosophical approach and analytic capabilities of a geographic information system to bear on a problem. A key area in the joint applications of remote sensing technology and GIS is to identify change. Whether this change is of interest for its own sake, or because the change causes us to act (for example, to update a map), remote sensing provides an excellent suite of tools for detecting change. At the same time, a GIS is perhaps the best analytic toot for quantifying the process of change. There are two alternative methods for quantifying changes. The conceptually simple approach is to un the pixel values in each of the images. This method is practical but may be too simple to identify the variety of changes in a complex scene. The common alternative is called symbolic change detection. The analyst first decides on a set of thematic categories that are important to distinguish for the application. This approach is useful only if accurate landuse/cover classifications can be obtained. Persons conducting digital change detection must be intimately familiar with the environment under study, the quality of the data set and the characteristics of change detection algorithms. Also, much work remains to identify optimum change detection algorithms for specific geographic areas and problems.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2007.04a
• /
• pp.389-393
• /
• 2007

There has been many researches using LiDAR(Light Detection And Ranging) data. There has been many other researches through out the world using the 3 dimensional spatial data in various fields. In this research, Using lidar data and digital images, we have extracted the position of the power-transmission line and created 3 dimensional models. The presented method is more efficient than field surveying and it can also be used lot monitoring change in the environment

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.37 no.3
• /
• pp.199-208
• /
• 2019

As the number of available satellites increases and technology advances, image information outputs are becoming increasingly diverse and a large amount of data is accumulating. In this study, we propose a change detection method for high-resolution satellite images that uses transfer learning and a deep learning network to overcome the limit caused by insufficient training data via the use of pre-trained information. The deep learning network used in this study comprises convolutional layers to extract the spatial and spectral information and convolutional long-short term memory layers to analyze the time series information. To use the learned information, the two initial convolutional layers of the change detection network are designed to use learned values from 40,000 patches of the ISPRS (International Society for Photogrammertry and Remote Sensing) dataset as initial values. In addition, 2D (2-Dimensional) and 3D (3-dimensional) kernels were used to find the optimized structure for the high-resolution satellite images. The experimental results for the KOMPSAT-3A (KOrean Multi-Purpose SATllite-3A) satellite images show that this change detection method can effectively extract changed/unchanged pixels but is less sensitive to changes due to shadow and relief displacements. In addition, the change detection accuracy of two sites was improved by using 3D kernels. This is because a 3D kernel can consider not only the spatial information but also the spectral information. This study indicates that we can effectively detect changes in high-resolution satellite images using the constructed image information and deep learning network. In future work, a pre-trained change detection network will be applied to newly obtained images to extend the scope of the application.

• Journal of Korean Society for Geospatial Information Science
• /
• v.21 no.3
• /
• pp.11-17
• /
• 2013

With high social demands for the diverse utilizations of forest lands, the illegal forest land use changes have increased. We studied change detection technique to detect changes in forest land use using an object-oriented segmentation of RED bands differencing in multi-temporal aerial photographs. The new object-oriented segmentation method consists of the 5 steps, "Image Composite - Segmentation - Reshaping - Noise Remover - Change Detection". The method enabled extraction of deforested objects by selecting a suitable threshold to determine whether the objects was divided or merged, based on the relations between the objects, spectral characteristics and contextual information from multi-temporal aerial photographs. The results found that the object-oriented segmentation method detected 12% of changes in forest land use, with 96% of the average detection accuracy compared by visual interpretation. Therefore this research showed that the spatial data by the object-oriented segmentation method can be complementary to the one by a visual interpretation method, and proved the possibility of automatically detecting and extracting changes in forest land use from multi-temporal aerial photographs.

• Proceedings of the Korea Contents Association Conference
• /
• 2008.05a
• /
• pp.897-901
• /
• 2008

The objective of this study is to conduct land cover classification respectively using Landsat TM data collected on Oct., 1985 and KOMPSAT-1 EOC data collected on Jan., 2000 covering Gumi city, Gyeongbuk Province and to detect urban change by comparing between both land cover maps. Multispectral images of Landsat TM have spatial resolution of 30m are well known as useful data for extracting information related to landcover, vegetation classification, urban growth analysis and so forth. In contrast, as KOMPSAT-1 EOC collects panchromatic images with relatively high spatial resolution of 6.6m. We try to analyze how accurate landcover classification result is able to be derived from the panchromatic images. As the results of the study, the KOMPSAT EOC data with high resolution greater than 4 times showed higher classification degree than Landsat TM data. It was ascertained that the built-up region was extended by three to four times in the last 15 years between 1985 and 2000. In the contrast, it was shown that the forest region was decreased by 15% to 27% and the grass region including agricultural region was decreased by 28% to 45%.

• Journal of Korean Society for Geospatial Information Science
• /
• v.15 no.4
• /
• pp.51-58
• /
• 2007

Information on forest type and cover density status of the present and past on large scale (1:50,000) is very much needed for conservation of any forest region. Such large-scale maps are not available for the Eastern Ghats (EG) of Tamil Nadu. This study deals with the preparation of forest type and cover density map of EG of Tamil Nadu during 2003 and the changes it has undergone between 1990 and 2003 using appropriate satellite data. About 10 forest types have been identified and mapped. Major changes have been observed in the forest types such as evergreen, and deciduous.

• Proceedings of the KSRS Conference
• /
• 2003.11a
• /
• pp.508-510
• /
• 2003

Land cover change has been recognized as one of the most important factors influencing the occurrence of rainfall-triggered landslides. Satellite remote sensing provides detailed information regarding the spatial distribution and extent of land cover/use changes. This study describes the land cover changes in Nagasaki City, Japan and Kangnung City, South Korea. The former has been suffered from rainfall-triggered disasters for long term and latter was damaged by Typhoon Rusa in 2002. The results obtained from both study areas clearly show that land cover changes have occurred in the last decade as a result of both natural forces and human activities.

• Journal of The Korean Society of Agricultural Engineers
• /
• v.60 no.5
• /
• pp.149-162
• /
• 2018

It is necessary to select the appropriate global climate model (GCM) to take into account the impacts of climate change on integrated water management. The objective of this study was to develop the selection technique of representative GCMs for uncertainty in climate change scenario. The selection technique which set priorities of GCMs consisted of two steps. First step was evaluating original GCMs by comparing with grid-based observational data for the past period. Second step was evaluating whether the statistical downscaled data reflect characteristics for the historical period. Spatial Disaggregation Quantile Delta Mapping (SDQDM), one of the statistical downscaling methods, was used for the downscaled data. The way of evaluating was using explanatory power, the stepwise ratio of the entire GCMs by Expert Team on Climate Change Detection and Indices (ETCCDI) basis. We used 26 GCMs based on CMIP5 data. The Representative Concentration Pathways (RCP) 4.5 and 8.5 scenarios were selected for this study. The period for evaluating reproducibility of historical period was 30 years from 1976 to 2005. Precipitation, maximum temperature, and minimum temperature were used as collected climate variables. As a result, we suggested representative 13 GCMs among 26 GCMs by using the selection technique developed in this research. Furthermore, this result can be utilized as a basic data for integrated water management.