• 한국항공우주학회지
• /
• 제49권10호
• /
• pp.883-892
• /
• 2021

본 논문에서는 드론을 활용한 감시정찰 임무의 효율성을 향상하기 위해 드론 탑재장비에서 실시간으로 구동 가능한 딥러닝 기반의 객체 인식 모델을 개발하는 연구를 수행하였다. 드론 영상 내 객체 인식 성능을 높이는 목적으로 학습 단계에서 학습 데이터 전처리 및 증강, 전이 학습을 수행하였고 각 클래스 별 성능 편차를 줄이기 위해 가중 크로스 엔트로피 방법을 적용하였다. 추론 속도를 개선하기 위해 양자화 기법이 적용된 추론 가속화 엔진을 생성하여 실시간성을 높였다. 마지막으로 모델의 성능을 확인하기 위해 학습에 참여하지 않은 드론 영상 데이터에서 인식 성능 및 실시간성을 분석하였다.

• 한국멀티미디어학회논문지
• /
• 제21권12호
• /
• pp.1425-1430
• /
• 2018

In this paper, we propose a CNN based deep learning algorithm for semantic segmentation of images. In order to improve the accuracy of semantic segmentation, we combined pixel level object classification and image level object classification. The image level object classification is used to accurately detect the characteristics of an image, and the pixel level object classification is used to indicate which object area is included in each pixel. The proposed network structure consists of three parts in total. A part for extracting the features of the image, a part for outputting the final result in the resolution size of the original image, and a part for performing the image level object classification. Loss functions exist for image level and pixel level classification, respectively. Image-level object classification uses KL-Divergence and pixel level object classification uses cross-entropy. In addition, it combines the layer of the resolution of the network extracting the features and the network of the resolution to secure the position information of the lost feature and the information of the boundary of the object due to the pooling operation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권9호
• /
• pp.4665-4683
• /
• 2019

Previous methods build image annotation model by leveraging three basic dependencies: relations between image and label (image/label), between images (image/image) and between labels (label/label). Even though plenty of researches show that multiple dependencies can work jointly to improve annotation performance, different dependencies actually do not "work jointly" in their diagram, whose performance is largely depending on the result predicted by image/label section. To address this problem, we propose the adaptive attention annotation model (AAAM) to associate these dependencies with the prediction path, which is composed of a series of labels (tags) in the order they are detected. In particular, we optimize the prediction path by detecting the relevant labels from the easy-to-detect to the hard-to-detect, which are found using Binary Cross-Entropy (BCE) and Triplet Margin (TM) losses, respectively. Besides, in order to capture the inforamtion of each label, instead of explicitly extracting regional featutres, we propose the self-attention machanism to implicitly enhance the relevant region and restrain those irrelevant. To validate the effective of the model, we conduct experiments on three well-known public datasets, COCO 2014, IAPR TC-12 and NUSWIDE, and achieve better performance than the state-of-the-art methods.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.153-166
• /
• 2022

Identifying fine cracks in steel bridge facilities is a challenging task of structural health monitoring (SHM). This study proposed an end-to-end crack image segmentation framework based on a one-step Convolutional Neural Network (CNN) for pixel-level object recognition with high accuracy. To particularly address the challenges arising from small object detection in complex background, efforts were made in loss function selection aiming at sample imbalance and module modification in order to improve the generalization ability on complicated images. Specifically, loss functions were compared among alternatives including the Binary Cross Entropy (BCE), Focal, Tversky and Dice loss, with the last three specialized for biased sample distribution. Structural modifications with dilated convolution, Spatial Pyramid Pooling (SPP) and Feature Pyramid Network (FPN) were also performed to form a new backbone termed CrackDet. Models of various loss functions and feature extraction modules were trained on crack images and tested on full-scale images collected on steel box girders. The CNN model incorporated the classic U-Net as its backbone, and Dice loss as its loss function achieved the highest mean Intersection-over-Union (mIoU) of 0.7571 on full-scale pictures. In contrast, the best performance on cropped crack images was achieved by integrating CrackDet with Dice loss at a mIoU of 0.7670.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.221-235
• /
• 2022

Manual inspection of steel box girders on long span bridges is time-consuming and labor-intensive. The quality of inspection relies on the subjective judgements of the inspectors. This study proposes an automated approach to detect and segment cracks in high-resolution images. An end-to-end cascaded framework is proposed to first detect the existence of cracks using a deep convolutional neural network (CNN) and then segment the crack using a modified U-Net encoder-decoder architecture. A Naïve Bayes data fusion scheme is proposed to reduce the false positives and false negatives effectively. To generate the binary crack mask, first, the original images are divided into 448 × 448 overlapping image patches where these image patches are classified as cracks versus non-cracks using a deep CNN. Next, a modified U-Net is trained from scratch using only the crack patches for segmentation. A customized loss function that consists of binary cross entropy loss and the Dice loss is introduced to enhance the segmentation performance. Additionally, a Naïve Bayes fusion strategy is employed to integrate the crack score maps from different overlapping crack patches and to decide whether a pixel is crack or not. Comprehensive experiments have demonstrated that the proposed approach achieves an 81.71% mean intersection over union (mIoU) score across 5 different training/test splits, which is 7.29% higher than the baseline reference implemented with the original U-Net.

• 한국수자원학회논문집
• /
• 제40권12호
• /
• pp.957-968
• /
• 2007

수자원의 정량적인 계획과 관리를 위해서는 정확하고 신뢰성 높은 유량 자료가 필수적이다. 이에 따라 최근에 초음파유량계와 유속지수법 등의 실시간 유량 측정 방법이 도입되고 있다. 이러한 방법들은 단면의 일부분에서 측정한 유속을 이용하여 전체 단면의 유량을 산정하고 있으므로 하천 단면의 2차원적 유속분포에 대한 합리적이고 이론적인 기초가 필요하다. 본 연구에서는 Chiu(1987, 1988)가 제안한 2차원 유속분포식을 자연하천에 적용하고 ADCP 실측 자료를 이용하여 비교 분석함으로써 적용성을 분석하였다. 이를 위해 실측 자료로부터 최대유속과 평균유속을 계산한 후 매개변수 M을 산정하였다. 등유속선 형상 매개변수는 최소자승합 기준의 목적함수를 이용하여 추정하였다. 최적화된 매개변수를 적용하여 도출된 엔트로피 유속분포를 실측 유속분포와 비교한 결과, 대체로 잘 일치하는 것으로 나타났다. 상관도가 높게 나타나는 14개의 실측 자료를 이용하여 매개변수 h, $\beta_i$의 특성을 분석한 후 미측정 단면에 적용할 수 있도록 그 값을 추정하였다. 추정된 매개변수를 검증을 위한 자료에 적용한 결과 역시 실측 자료를 대체로 잘 재현하는 것으로 나타났다. 유량의 경우 최대 7% 의 오차로 실측 자료와 대체로 비슷하게 산정하였다. Chiu의 유속분포식에 관여하는 매개변수를 적절히 추정한다면 자연하천의 유속분포를 잘 모의할 수 있을 것으로 판단된다.

• Structural Engineering and Mechanics
• /
• 제58권3호
• /
• pp.459-473
• /
• 2016

Various vague and unstructured problems encountered the civil engineering/designers that persuaded by their experiences. One of these problems is the structural failure of the reinforced concrete (RC) building determination. Typically, using the traditional Limit state method is time consuming and complex in designing structures that are optimized in terms of one/many parameters. Recent research has revealed the Artificial Neural Networks potentiality in solving various real life problems. Thus, the current work employed the Multilayer Perceptron Feed-Forward Network (MLP-FFN) classifier to tackle the problem of predicting structural failure of multistoried reinforced concrete buildings via detecting the failure possibility of the multistoried RC building structure in the future. In order to evaluate the proposed method performance, a database of 257 multistoried buildings RC structures has been constructed by professional engineers, from which 150 RC structures were used. From the structural design, fifteen features have been extracted, where nine features of them have been selected to perform the classification process. Various performance measures have been calculated to evaluate the proposed model. The experimental results established satisfactory performance of the proposed model.

• 한국전자통신학회논문지
• /
• 제10권6호
• /
• pp.683-690
• /
• 2015

러프집합에서 누락된 속성 값들은 Reduct와 Core 계산, 더 나아가서 결정 트리 구축에 있어서 식별 불능의 패턴 부합 문제를 가진다. 현재 누락된 속성 값들의 추정과 관련하여 보편적인 속성 값으로의 대체, 속성들의 모든 가능한 값 할당, 이벤트 포장 방법, C4.5, 특수한 LEM2 알고리즘과 같은 접근방식들이 적용되고 있다. 그렇지만, 이들 접근방식은 결국 전형적으로 자주 등장하는 속성 값 혹은 가장 보편적인 속성 값으로의 단순 대체를 나타내기 때문에, 주요 속성 값들이 누락된 경우에 정보 손실이 큰 의사 결정 규칙들이 유도되기 때문에 의사결정 규칙들의 교차 검증에서 문제가 된다. 본 연구에서는 이러한 문제점을 개선시키기 위해 속성들간에 엔트로피 변동을 활용하여 정보 이득이 높은 방향으로 누락된 속성 값들을 대체하는 방식을 제안한다. 제안된 접근방식에 관한 타당성 검토는 비교적 가까운 유사 관계에 의해 누락 값 대체 방식을 적용하는 ROSE 프로그램과의 비교를 나타낸다.

• 한국수자원학회논문집
• /
• 제41권6호
• /
• pp.575-585
• /
• 2008

수자원의 계획 평가 관리 및 수공구조물의 설계를 위해서는 정확하고 신뢰성 높은 유량 자료가 필수적이다. 본 연구에서는 Chiu의 유속분포와 최대유속 추정을 이용하여 하천유량을 계산하는 새로운 방법을 제시하였다. 기존 면적유속법과 비교 검토한 바, 본 연구에서 개발한 방법은 기존 유속면적법과 매우 유사한 하천유량을 보였다. Price-AA를 이용하여 유속을 측정할 경우 측선의 수심에 따라 정해진 지점에서 유속을 측정하여야 하는데, 본 연구에서 제시한 방법을 이용하면 임의의 측선과 측점에서 유속을 측정하여도 정확한 유량계산이 가능하다. 그러나 흐름 단면이 매우 복잡하거나 좌우의 비대칭성이 심한 경우에는 엔트로피 개념의 Chiu의 유속분포가 실제 자연하천의 흐름분포에서 멀어지고 유량산정에 Chiu의 유속분포의 정확도가 떨어지기 때문에 본 연구에서 제시한 방법을 적용하기 어렵다.

• 환경영향평가
• /
• 제21권4호
• /
• pp.593-607
• /
• 2012

The Ministry of Environment have started the 'National Ecosystem Survey' since 1986. It has been carried out nationwide every ten years as the largest survey project in Korea. The second one and the third one produced the GIS-based inventory of species. Three survey methods were different from each other. There were few studies for species distribution using national survey data in Korea. The purposes of this study are to test species distribution models for finding the most suitable modeling methods for the National Ecosystem Survey data and to investigate the modeling results according to survey methods and taxonominal group. Occurrence data of nine species were extracted from the National Ecosystem Survey by taxonomical group (plant, mammal, and bird). Plants are Korean winter hazel (Corylopsis coreana), Iris odaesanensis (Iris odaesanensis), and Berchemia (Berchemia berchemiaefolia). Mammals are Korean Goral (Nemorhaedus goral), Marten (Martes flavigula koreana), and Leopard cat (Felis bengalensis). Birds are Black Woodpecker (Dryocopus martius), Eagle Owl (Bubo Bubo), and Common Buzzard (Buteo buteo). Environmental variables consisted of climate, topography, soil and vegetation structure. Two modeling methods (GAM, Maxent) were tested across nine species, and predictive species maps of target species were produced. The results of this study were as follows. Firstly, Maxent showed similar 5 cross-validated AUC with GAM. Maxent is more useful model to develop than GAM because National Ecosystem Survey data has presence-only data. Therefore, Maxent is more useful species distribution model for National Ecosystem Survey data. Secondly, the modeling results between the second and third survey methods showed sometimes different because of each different surveying methods. Therefore, we need to combine two data for producing a reasonable result. Lastly, modeling result showed different predicted distribution pattern by taxonominal group. These results should be considered if we want to develop a species distribution model using the National Ecosystem Survey and apply it to a nationwide biodiversity research.