• Korean Journal of Agricultural and Forest Meteorology
• /
• v.20 no.2
• /
• pp.205-213
• /
• 2018

Development and validation of crop models often require measurements of biomass for the crop of interest. Considerable efforts would be needed to obtain a reasonable amount of biomass data because the destructive sampling of a given crop is usually used. The Kinect sensor, which has a combination of image and depth sensors, can be used for estimating crop biomass without using destructive sampling approach. This approach could provide more data sets for model development and validation. The objective of this study was to examine the applicability of the Kinect sensor for estimation of chinese cabbage fresh weight. The fresh weight of five chinese cabbage was measured and compared with estimates using the Kinect sensor. The estimates were obtained by scanning individual chinese cabbage to create point cloud, removing noise, and building a three dimensional model with a set of free software. It was found that the 3D model created using the Kinect sensor explained about 98.7% of variation in fresh weight of chinese cabbage. Furthermore, the correlation coefficient between estimates and measurements were highly significant, which suggested that the Kinect sensor would be applicable to estimation of fresh weight for chinese cabbage. Our results demonstrated that a depth sensor allows for a non-destructive sampling approach, which enables to collect observation data for crop fresh weight over time. This would help development and validation of a crop model using a large number of reliable data sets, which merits further studies on application of various depth sensors to crop dry weight measurements.

• Journal of Broadcast Engineering
• /
• v.24 no.6
• /
• pp.1076-1084
• /
• 2019

In this paper, we introduced a system that extracts metadata by recognizing characters and objects in media using deep learning technology. In the field of broadcasting, multimedia contents such as video, audio, image, and text have been converted to digital contents for a long time, but the unconverted resources still remain vast. Building media archives requires a lot of manual work, which is time consuming and costly. Therefore, by implementing a deep learning-based metadata generation system, it is possible to save time and cost in constructing media archives. The whole system consists of four elements: training data generation module, object recognition module, character recognition module, and API server. The deep learning network module and the face recognition module are implemented to recognize characters and objects from the media and describe them as metadata. The training data generation module was designed separately to facilitate the construction of data for training neural network, and the functions of face recognition and object recognition were configured as an API server. We trained the two neural-networks using 1500 persons and 80 kinds of object data and confirmed that the accuracy is 98% in the character test data and 42% in the object data.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1663-1676
• /
• 2022

Recently, a number of deep-learning based land cover segmentation studies have been introduced. Some studies denoted that the performance of land cover segmentation deteriorated due to insufficient training data. In this study, we verified the improvement of land cover segmentation performance through data augmentation. U-Net was implemented for the segmentation model. And 2020 satellite-derived landcover dataset was utilized for the study data. The pixel accuracies were 0.905 and 0.923 for U-Net trained by original and augmented data respectively. And the mean F1 scores of those models were 0.720 and 0.775 respectively, indicating the better performance of data augmentation. In addition, F1 scores for building, road, paddy field, upland field, forest, and unclassified area class were 0.770, 0.568, 0.433, 0.455, 0.964, and 0.830 for the U-Net trained by original data. It is verified that data augmentation is effective in that the F1 scores of every class were improved to 0.838, 0.660, 0.791, 0.530, 0.969, and 0.860 respectively. Although, we applied data augmentation without considering class balances, we find that data augmentation can mitigate biased segmentation performance caused by data imbalance problems from the comparisons between the performances of two models. It is expected that this study would help to prove the importance and effectiveness of data augmentation in various image processing fields.

• Spatial Information Research
• /
• v.8 no.1
• /
• pp.141-153
• /
• 2000

The purpose of this study is to develop and apply spatial landslide information system using Geographic information system (GIS) in concerned with spatial data. Landslide locations detected from interpretation of aerial photo and field survey, and topographic , soil , forest , and geological maps of the study area, Yongin were collected and constructed into spatial database using GIS. As landslide occurrence factors, slope, aspect and curvature of topography were calculated from the topographic database. Texture, material, drainage and effective thickness of soil were extracted from the soil database, and type, age, diameter and density of wood were extracted from the forest database. Lithology was extracted from the geological database, and land use was classified from the Landsat TM satellite image. In addition, landslide damageable objects such as building, road, rail and other facility were extracted from the topographic database. Landslide susceptibility was analyzed using the landslide occurrence factors by probability, logistic regression and neural network methods. The spatial landslide information system was developed to retrieve the constructed GIS database and landslide susceptibility . The system was developed using Arc View script language(Avenue), and consisted of pull-down and icon menus for easy use. Also, the constructed database can be retrieved through Internet World Wide Web (WWW) using Internet GIS technology.

• Korean Journal of Optics and Photonics
• /
• v.29 no.5
• /
• pp.204-214
• /
• 2018

Since LIDAR is advantageous for accurate information acquisition and realization of a high-resolution 3D image based on characteristics that can be precisely measured, it is essential to autonomous navigation systems that require acquisition and judgment of accurate peripheral information without user intervention. Recently, as an autonomous navigation system applying LIDAR has been utilized in human living space, it is necessary to solve the eye-safety problem, and to make reliable judgment through accurate obstacle recognition in various environments. In this paper, we construct a single-shot LIDAR system (SSLs) using a 1550-nm eye-safe light source, and report the analysis method and results of LIDAR signals for various measurement environments, reflective materials, and material angles. We analyze the signals of materials with different reflectance in each measurement environment by using a 5% Al reflector and a building wall located at a distance of 25 m, under indoor, daytime, and nighttime conditions. In addition, signal analysis of the angle change of the material is carried out, considering actual obstacles at various angles. This signal analysis has the merit of possibly confirming the correlation between measurement environment, reflection conditions, and LIDAR signal, by using the SNR to determine the reliability of the received information, and the timing jitter, which is an index of the accuracy of the distance information.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.37 no.3
• /
• pp.69-81
• /
• 2009

This study was intended to evaluate the objective landscape of rural region using an additive integration index method in the Seondong region of Gochang-gun, Jeollabuk-do, Korea. This study consisted of the following three steps. First, this study developed an additive integration index calculation model for landscape assessment based on indicators and weight to each space type in accordance with three landscape fields which were developed by the expert Delphi method. Second, this study used NDVI (Normalized Difference Vegetation Index) and permeable area rate, which were available from high resolution satellite image, to calculate the green naturality degree, area rate, and building coverage respectively. Third, this study has calculated the landscape assessment index of rural regions using an additive integration index method made of assessment data and weight for each indicator. This study has found the following results: 1) landscape level was very poor in all 6 types of space, marking grade five; 2) while the highest level of natural landscape and mixed landscape was grade two, that of artificial landscape was grade five; 3) based on objective landscape, grade five showed the highest frequency, and grade one, two, three, and four followed in that order.

• Archives of design research
• /
• v.14 no.2
• /
• pp.117-126
• /
• 2001

Dietary life is said to be consummated even to a phase of art from phase of survival, recognition, selection and preference. Dietary life, from a sheer earing, is now transforming itself into a joy, and therefore, function of foodstuff packaging demand diversification rather than a simple packaging. Recently, socio-economic environment is showing conspicuous changes in that there are increasing number of working couples nuclear family, old-age population, social activity and improvement of living standards. Such a change in the living environment has much impact on our dietary life. Particularly, an age when cake and cookies of western origin were all but strange is gradually phasing out, while it is deemed to be improper to overlook the fact that bread and cakes are solidifying their position as part of foodstuff meanies for on dietary life. Of cakes and cookies, cakes have come to enjoy a position whereby they are regarded as part of an imperative for family banquets, various get-together and birthday celebration, etc., with a significant improvement that caters to our taste vis-a-vis an stage of introduction of cakes. At the time when reined foreign bakeries, one after another , are contemplating to make an inroad into Korea, and when the distribution market is to be opened fully in July of 1993, Korea bakeries that have been building up their position within the domestic market are expected to face a considerable number of difficulties. Accordingly, under such circumstances of the time text we are in it is attempted in this study to map out measures that may contribute to the strengthening of the products of business enterprises and improvement of corporate image that may appeal to feeling and emotions of consumers packaging that could attain objectives, and package design planning that, as an important factor for playing its own share in the restoration of humanism that is being alienated, may appeal to consumer sensitivity, rather than packaging that is being utilized merely as tool designed for marketing

• Journal of Cadastre & Land InformatiX
• /
• v.50 no.1
• /
• pp.75-85
• /
• 2020

This study examines machine learning cross-validation to utilized create ROI for classification of land cover. The study area located in Sejong and one KOMPSAT-3A image was used in this analysis: procedure on October 28, 2019. We used four bands(Red, Green, Blue, Near infra-red) for learning cross validation process. In this study, we used K-fold method in cross validation and used SVM kernel type with cross validation result. In addition, we used 4 kernels of SVM(Linear, Polynomial, RBF, Sigmoid) for supervised classification land cover map using extracted ROI. During the cross validation process, 1,813 data extracted from 3,500 data, and the most of the building, road and grass class data were removed about 60% during cross validation process. Based on this, the supervised SVM linear technique showed the highest classification accuracy of 91.77% compared to other kernel methods. The grass' producer accuracy showed 79.43% and identified a large mis-classification in forests. Depending on the results of the study, extraction ROI using cross validation may be effective in forest, water and agriculture areas, but it is deemed necessary to improve the distinction of built-up, grass and bare-soil area.

• Proceedings of the KSRS Conference
• /
• v.2
• /
• pp.551-554
• /
• 2006

House management is very important in water resource protection in order to provide sustainable drinking water for about four millions population in northern Taiwan. House management can be a simple job that can be done without any ingredient of remote sensing or geographic information systems. Remote sensing and GIS integration for house management can provide more efficient management prescription when land use enforcement, soil and water conservation, sewage management, garbage collection, and reforestation have to be managed simultaneously. The objective of this paper was to integrate remote sensing and GIS to manage houses in a water resource protection district. More than four thousand houses have been surveyed and created as a house data base. Site map of every single house and very detail information consisting of address, ownership, date of creation, building materials, acreages floor by floor, parcel information, and types of house condition. Some houses have their photos in different directions. One house has its own card consists these information and these attributes were created into a house data base. Site maps of all houses were created with the same coordinates system as parcel maps, topographic maps, sewage maps, and city planning maps. Visual Basic.NET, Visual C#.NET have been implemented to develop computer programs for house information inquiry and maps overlay among house maps and other GIS map layers. Remote sensing techniques have been implemented to generate the background information of a single house in the past 15 years. Digital orthophoto maps at a scale of 1:5000 overlay with house site maps are very useful in determination of a house was there or not for a given year. Satellite images if their resolutions good enough are also very useful in this type of daily government operations. The developed house management systems can work with commercial GIS software such as ArcView and ArcPad. Remote sensing provided image information of a single house whether it was there or not in a given year. GIS provided overlay and inquiry functions to automatically extract attributes of a given house by ownership, address, and so on when certain house management prescriptions have to be made by government agency. File format is the key component that makes remote sensing and GIS integration smoothly. The developed house management systems are user friendly and can be modified to meet needs encountered in a single task of a government technician.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.20 no.2
• /
• pp.95-111
• /
• 2021

As the population decreases in an aging society, the average age of drivers increases. Accordingly, the elderly at high risk of being in an accident need autonomous-driving vehicles. In order to secure driving safety on the road, several technologies to respond to various obstacles are required in those vehicles. Among them, technology is required to recognize static obstacles, such as poor road conditions, as well as dynamic obstacles, such as vehicles, bicycles, and people, that may be encountered while driving. In this study, we propose a deep neural network algorithm capable of simultaneously detecting these two types of obstacle. For this algorithm, we used 1,418 road images and produced annotation data that marks seven categories of dynamic obstacles and labels images to indicate road damage. As a result of training, dynamic obstacles were detected with an average accuracy of 46.22%, and road surface damage was detected with a mean intersection over union of 74.71%. In addition, the average elapsed time required to process a single image is 89ms, and this algorithm is suitable for personal mobility vehicles that are slower than ordinary vehicles. In the future, it is expected that driving safety with personal mobility vehicles will be improved by utilizing technology that detects road obstacles.