• Journal of Broadcast Engineering
• /
• v.27 no.4
• /
• pp.487-498
• /
• 2022

Depth map is an image that contains distance information in 3D space on a 2D plane and is used in various 3D vision tasks. Many existing depth estimation studies mainly use narrow FoV images, in which a significant portion of the entire scene is lost. In this paper, we propose a technique for generating 360° omnidirectional RGBD images from a sparse set of narrow FoV images. The proposed generative adversarial network based image generation model estimates the relative FoV for the entire panoramic image from a small number of non-overlapping images and produces a 360° RGB and depth image simultaneously. In addition, it shows improved performance by configuring a network reflecting the spherical characteristics of the 360° image.

• Investigative Magnetic Resonance Imaging
• /
• v.16 no.1
• /
• pp.6-15
• /
• 2012

Purpose : The objective of this study was to develop background gradient correction method using excitation pulse profile compensation for accurate fat and $T_2{^*}$ quantification in the liver. Materials and Methods: In liver imaging using gradient echo, signal decay induced by linear background gradient is weighted by an excitation pulse profile and therefore hinders accurate quantification of $T_2{^*}$and fat. To correct this, a linear background gradient in the slice-selection direction was estimated from a $B_0$ field map and signal decays were corrected using the excitation pulse profile. Improved estimation of fat fraction and $T_2{^*}$ from the corrected data were demonstrated by phantom and in vivo experiments at 3 Tesla magnetic field. Results: After correction, in the phantom experiments, the estimated $T_2{^*}$ and fat fractions were changed close to that of a well-shimmed condition while, for in vivo experiments, the background gradients were estimated to be up to approximately 120 ${\mu}T/m$ with increased homogeneity in $T_2{^*}$ and fat fractions obtained. Conclusion: The background gradient correction method using excitation pulse profile can reduce the effect of macroscopic field inhomogeneity in signal decay and can be applied for simultaneous fat and iron quantification in 2D gradient echo liver imaging.

• Journal of Internet Computing and Services
• /
• v.18 no.6
• /
• pp.85-92
• /
• 2017

Recently, in the field of video surveillance, a Deep Learning based learning method has been applied to a method of detecting a moving person in a video and analyzing the behavior of a detected person. The human activity recognition, which is one of the fields this intelligent image analysis technology, detects the object and goes through the process of detecting the body keypoint to recognize the behavior of the detected object. In this paper, we propose a method for Body Keypoint Localization based on Object Detection using RGB-D information. First, the moving object is segmented and detected from the background using color information and depth information generated by the two cameras. The input image generated by rescaling the detected object region using RGB-D information is applied to Convolutional Pose Machines for one person's pose estimation. CPM are used to generate Belief Maps for 14 body parts per person and to detect body keypoints based on Belief Maps. This method provides an accurate region for objects to detect keypoints an can be extended from single Body Keypoint Localization to multiple Body Keypoint Localization through the integration of individual Body Keypoint Localization. In the future, it is possible to generate a model for human pose estimation using the detected keypoints and contribute to the field of human activity recognition.

• Journal of Korean Society of Forest Science
• /
• v.111 no.3
• /
• pp.435-449
• /
• 2022

Forest canopy height is an indispensable vertical structure parameter that can be used for understanding forest biomass and carbon storage as well as for managing a sustainable forest ecosystem. Plot-based field surveys, such as the national forest inventory, have been conducted to provide estimates of the forest canopy height. However, the comprehensive nationwide field monitoring of forest canopy height has been limited by its cost, lack of spatial coverage, and the inaccessibility of some forested areas. These issues can be addressed by remote sensing technology, which has gained popularity as a means to obtain detailed 2- and 3-dimensional measurements of the structure of the canopy at multiple scales. Here, we reviewed both international and domestic studies that have used remote sensing technology approaches to estimate the forest canopy height. We categorized and examined previous approaches as: 1) LiDAR approach, 2) Stereo or SAR image-based point clouds approach, and 3) combination approach of remote sensing data. We also reviewed upscaling approaches of utilizing remote sensing data to generate a continuous map of canopy height across large areas. Finally, we provided suggestions for further advancement of the Korean forest canopy height estimation system through the use of various remote sensing technologies.

• Atmosphere
• /
• v.33 no.1
• /
• pp.61-72
• /
• 2023

Accurate estimation of forest carbon stocks is important in establishing greenhouse gas reduction plans. In this study, we estimate the spatial distribution of forest carbon stocks using machine learning techniques based on high-resolution remote sensing data and detailed field survey data. The high-resolution remote sensing data used in this study are Landsat indices (EVI, NDVI, NDII) for monitoring vegetation vitality and Shuttle Radar Topography Mission (SRTM) data for describing topography. We also used the forest growing stock data from the National Forest Inventory (NFI) for estimating forest biomass. Based on these data, we built a model based on machine learning methods and optimized for Korean forest types to calculate the forest carbon stocks per grid unit. With the newly developed estimation model, we created forest carbon stocks maps and estimated the forest carbon stocks in South Korea. As a result, forest carbon stock in South Korea was estimated to be 432,214,520 tC in 2020. Furthermore, we estimated the loss of forest carbon stocks due to the Donghae-Uljin forest fire in 2022 using the forest carbon stock map in this study. The surrounding forest destroyed around the fire area was estimated to be about 24,835 ha and the loss of forest carbon stocks was estimated to be 1,396,457 tC. Our model serves as a tool to estimate spatially distributed local forest carbon stocks and facilitates accounting of real-time changes in the carbon balance as well as managing the LULUCF part of greenhouse gas inventories.

• Korean Journal of Remote Sensing
• /
• v.25 no.4
• /
• pp.311-320
• /
• 2009

This study was carried out to estimate forest biomass and to produce forest biomass thematic map for Muju county by combining field data from the 5$^{th}$ National Forest Inventory (2006-2007) and satellite data. For estimating forest biomass, two methods were examined using a Landsat TM-5(taken on April 28th, 2005) and field data: multi-variant regression modeling and t-Nearest Neighbor (k-NN) technique. Estimates of forest biomass by the two methods were compared by a cross-validation technique. The results showed that the two methods provide comparatively accurate estimation with similar RMSE (63.75$\sim$67.26ton/ha) and mean bias ($\pm$1ton/ha). However, it is concluded that the k-NN method for estimating forest biomass is superior in terms of estimation efficiency to the regression model. The total forest biomass of the study site is estimated 8.4 million ton, or 149 ton/ha by the k-NN technique.

• Journal of Soil and Groundwater Environment
• /
• v.20 no.7
• /
• pp.23-33
• /
• 2015

This study was conducted to provide the preliminary data on preparation for policy decisions regarding the groundwater supply scheme for N. Korea vulnerable to drought. Agricultural activities oriented to upland field due to the mountainous terrain more than 79% as well as the diversity of rainfall distribution over the country make more vulnerable to structural problems in the drought in N. Korea. Therefore, in anticipation of the expansion of exchange policies in agricultural sector, the available amount of groundwater needed for sustainable water resources supply was estimated for each administrative district after analyzing alluvium and hydrogeology distribution in N. Korea. Overall, the available amount of groundwater was estimated to 22.3 billion m³ (0.18 million m³/km²). The available amount of groundwater per unit area in Ryanggangdo and Gaesungsi was appeared very high in each of 0.56 and 0.39 million m³/km², respectively, and it would be interpreted that two districts have relatively wide area of volcanic rocks and alluvium with highly permeable characteristics, respectively. Finally, to maximize the utilization of this study result, the available amount of groundwater distribution map was developed on the basis of the 1 × 1 km grid network over the entire N. Korea.

• Journal of Wetlands Research
• /
• v.14 no.3
• /
• pp.317-328
• /
• 2012

The objects of this study are to perform appropriateness analysis of USLE(universal soil loss equation) model and to accumulate the data measured in field. The basin area of Bongdong station is $342.27km^2$. This study simulated sediment outflows in the basin and performed a comparative analysis of simulated outputs with actual measurement values. Also annual rainfall was used to calculate rainfall-runoff erosivity factor which can influence soil erosion. The calculation of annual average soil erosion was made by soil erosion maps. The maps with a resolution of ($30m{\times}30m$) were created by multiplication of factors(R, LS, K, C, P) from ArcView Map Calculator. In this paper, it was shown that soil erosion was not occur in the most of basin.

• Journal of Environmental Science International
• /
• v.25 no.8
• /
• pp.1165-1176
• /
• 2016

The accuracy of digital elevation models (DEMs) is crucial for properly estimating flood inundation area. DEM pixel size is especially important when generating flood inundation maps of small streams with a channel width of less than 50 m. In Korea, DEMs with large spatial resolutions of 30 m have been widely applied to generate flood inundation maps, even for small streams. Additionally, when making river master plans, field observations of stream cross-sections, as well as reference points in the middle of the river, have not previously been used to enhance the DEM. In this study, it was graphically demonstrated that high-resolution DEMs can increase the accuracy of flood inundation mapping, especially for small streams. Also, a methodology was proposed to modify the existing low-resolution DEMs by adding additional survey reference points, including river cross-sections, and interpolating them into a high spatial resolution DEM using the inverse distance weighting method. For verification purposes, the modified DEM was applied to Han stream on Jeju Island. The modified DEM showed much better accuracy when describing morphological features near the stream. Moreover, the flood inundation maps were formulated with the original 30 m pixel DEM and the modified 0.1 m pixel DEM using HEC-RAS modeling of the actual flood event of Typhoon Nari, and then compared with the flood history map of Nari. The results clearly indicated that the modified DEM generated a similar inundation area, but a very poor estimate of inundation area was derived from the original low-resolution DEM.

• Journal of Korea Water Resources Association
• /
• v.36 no.4
• /
• pp.633-645
• /
• 2003

The techniques of GIS and remote sensing are being applied to hydrology, geomorphology and various field of studies are performed by many researcher, related those techniques. In this paper, curve number change detection is tested according to soil map and land cover in mountain area. Neural networks method is applied for land cover classification and GIS for curve number calculation. The first, sample area are selected and tested land cover classification, NN(84.1％) is superior to MLC(80.9％). So we selected NN with land cover classifier. The second, curve number from the land cover by neural network classifier(57) is compared with that(curve number) from the land cover by manual work(55). Two values are so similar. The third, curve number classified by NN in sample area was applied and tested to whole study area. As results of this study, it is shown that curve number is more exact and efficient by using NN and GIS technique than by (using) manual work.