• Journal of Internet Computing and Services
• /
• v.21 no.6
• /
• pp.23-31
• /
• 2020

This study was carried out to generate various images of railroad surfaces with random defects as training data to be better at the detection of defects. Defects on the surface of railroads are caused by various factors such as friction between track binding devices and adjacent tracks and can cause accidents such as broken rails, so railroad maintenance for defects is necessary. Therefore, various researches on defect detection and inspection using image processing or machine learning on railway surface images have been conducted to automate railroad inspection and to reduce railroad maintenance costs. In general, the performance of the image processing analysis method and machine learning technology is affected by the quantity and quality of data. For this reason, some researches require specific devices or vehicles to acquire images of the track surface at regular intervals to obtain a database of various railway surface images. On the contrary, in this study, in order to reduce and improve the operating cost of image acquisition, we constructed the 'Defective Railroad Surface Regeneration Model' by applying the methods presented in the related studies of the Generative Adversarial Network (GAN). Thus, we aimed to detect defects on railroad surface even without a dedicated database. This constructed model is designed to learn to generate the railroad surface combining the different railroad surface textures and the original surface, considering the ground truth of the railroad defects. The generated images of the railroad surface were used as training data in defect detection network, which is based on Fully Convolutional Network (FCN). To validate its performance, we clustered and divided the railroad data into three subsets, one subset as original railroad texture images and the remaining two subsets as another railroad surface texture images. In the first experiment, we used only original texture images for training sets in the defect detection model. And in the second experiment, we trained the generated images that were generated by combining the original images with a few railroad textures of the other images. Each defect detection model was evaluated in terms of 'intersection of union(IoU)' and F1-score measures with ground truths. As a result, the scores increased by about 10~15% when the generated images were used, compared to the case that only the original images were used. This proves that it is possible to detect defects by using the existing data and a few different texture images, even for the railroad surface images in which dedicated training database is not constructed.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1357-1369
• /
• 2022

Cloud removal is an essential image processing step for any task requiring time-series optical images, such as vegetation monitoring and change detection. This paper presents a two-stage cloud removal method that combines conditional generative adversarial networks (cGANs) with regression-based calibration to construct a cloud-free time-series optical image set. In the first stage, the cGANs generate initial prediction results using quantitative relationships between optical and synthetic aperture radar images. In the second stage, the relationships between the predicted results and the actual values in non-cloud areas are first quantified via random forest-based regression modeling and then used to calibrate the cGAN-based prediction results. The potential of the proposed method was evaluated from a cloud removal experiment using Sentinel-2 and COSMO-SkyMed images in the rice field cultivation area of Gimje. The cGAN model could effectively predict the reflectance values in the cloud-contaminated rice fields where severe changes in physical surface conditions happened. Moreover, the regression-based calibration in the second stage could improve the prediction accuracy, compared with a regression-based cloud removal method using a supplementary image that is temporally distant from the target image. These experimental results indicate that the proposed method can be effectively applied to restore cloud-contaminated areas when cloud-free optical images are unavailable for environmental monitoring.

• Journal of Korean Society for Geospatial Information Science
• /
• v.18 no.4
• /
• pp.61-69
• /
• 2010

The collapse of a slope or cut embankment brings much damage to life and property. Accordingly, it is very important to analyze the spatial distribution by calculating the landslide susceptibility in the estimation of the risk of landslide occurrence. The heuristic, statistic, deterministic, and probabilistic methods have been introduced to make landslide susceptibility maps. In many cases, however, the reliability is low due to insufficient field data, and the qualitative experience and knowledge of experts could not be combined with the quantitative mechanical?analysis model in the existing methods. In this paper, new modeling method for a probabilistic landslide susceptibility analysis combined Bayesian Network with ontology model about experts' knowledge and spatial data was proposed. The ontology model, which was made using the reasoning engine, was automatically converted into the Bayesian Network structure. Through conditional probabilistic reasoning using the created Bayesian Network, landslide susceptibility with uncertainty was analyzed, and the results were described in maps, using GIS. The developed Bayesian Network was then applied to the test-site to verify its effect, and the result corresponded to the landslide traces boundary at 86.5% accuracy. We expect that general users will be able to make a landslide susceptibility analysis over a wide area without experts' help.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.727-729
• /
• 2005

나이브베이즈분류기($na\ddot{i}ve$ Bayes classifier)는 학습, 적용 및 계산자원 이용의 측면에서 매우 효율적인 모델이다. 또한, 그 분류 성능 역시 다른 기법에 비해 크게 떨어지지 않음이 다양한 실험을 통해 보여져 왔다. 특히, 데이터를 생성한 실제 확률분포를 나이브베이즈분류기가 정확하게 표현할 수 있는 경우에는 최대의 효과를 볼 수 있다. 하지만, 실제 확률분포에 존재하는 조건부독립성(conditional independence)이 나이브베이즈분류기의 구조와 일치하지 않는 경우에는 성능이 하락할 수 있다. 보다 구체적으로, 각 자질변수(feature variable)들 사이에 확률적 의존관계(probabilistic dependency)가 존재하는 경우 성능 하락은 심화된다. 본 논문에서는 이러한 나이브베이즈분류기의 약점을 효율적으로 해결할 수 있는 자질변수의 통합기법을 제시한다. 자질변수의 통합은 각 변수들 사이의 관계를 명시적으로 표현해 주는 방법이며, 특히 상호정보량(mutual information)에 기반한 통합 변수의 선정이 성능 향상에 크게 기여함을 실험을 통해 보인다.

• The KIPS Transactions:PartB
• /
• v.10B no.6
• /
• pp.611-618
• /
• 2003

In this paper, we describe a fuzzy inference approach for detecting and classifying shot transitions in video sequences. Our approach basically extends FAM (Fuzzy Associative Memory) to detect and classify shot transitions, including cuts, fades and dissolves. We consider a set of feature values that characterize differences between two consecutive frames as input fuzzy sets, and the types of shot transitions as output fuzzy sets. The inference system proposed in this paper is mainly composed of a learning phase and an inferring phase. In the learning phase, the system initializes its basic structure by determining fuzzy membership functions and constructs fuzzy rules. In the inferring phase, the system conducts actual inference using the constructed fuzzy rules. In order to verify the performance of the proposed shot transition detection method experiments have been carried out with a video database that includes news, movies, advertisements, documentaries and music videos.

• Journal of KIISE:Computer Systems and Theory
• /
• v.36 no.4
• /
• pp.274-282
• /
• 2009

Various location tracking sensors; such as GPS, INS, radar, and optical equipment; are used for tracking moving targets. In order to effectively track moving targets, it is necessary to develop an effective fusion method for these heterogeneous devices. There have been studies in which the estimated values of each sensors were regarded as different models and fused together, considering the different error characteristics of the sensors for the improvement of tracking performance using heterogeneous multi-sensor. However, the rate of errors for the estimated values of other sensors has increased, in that there has been a sharp increase in sensor errors and the attempts to change the estimated sensor values for the Sensor Probability could not be applied in real time. In this study, the Sensor Probability is obtained by comparing the RMSE (Root Mean Square Error) for the difference between the updated and measured values of the Kalman filter for each sensor. The process of substituting the new combined values for the Kalman filter input values for each sensor is excluded. There are improvements in both the real-time application of estimated sensor values, and the tracking performance for the areas in which the sensor performance has rapidly decreased. The proposed algorithm adds the error characteristic of each sensor as a conditional probability value, and ensures greater accuracy by performing the track fusion with the sensors with the most reliable performance. The trajectory of a UAV is generated in an experiment and a performance analysis is conducted with other fusion algorithms.