• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2021년도 제64차 하계학술대회논문집 29권2호
• /
• pp.51-52
• /
• 2021

본 논문에서는 얼굴 검증(Face Verification) 문제를 해결하기 위한 방법론으로 깊은 삼중 항 네트워크 모델을 제안한다. 본 논문에서는 얼굴 검증을 거리기반 유사도 문제로 보고, 딥러닝 기반 메트릭 러닝으로 해결하고자 하였다. 딥 메트릭 러닝 중 하나인 삼중 항 네트워크를 깊게 쌓기 위해 ResNet50, ResNet101과 경량화 모델인 MobileNet v3를 적용하였으며, 위 모델을 사용함으로써 이미지의 특징 추출을 효과적으로 할 수 있었다. 본 연구에서 제시한 방법론은 추후 복잡한 모델이 필요한 영상 데이터 내 얼굴 식별 모델에 기초 연구로서의 의의가 있다.

• 한국컴퓨터정보학회논문지
• /
• 제27권9호
• /
• pp.1-12
• /
• 2022

안과 환자의 질병을 판단하기 위해서는 특수 촬영 장비를 통해 찍은 안구영상을 이용한 안과의사의 주관적 판단의 개입이 전통적으로 활용되고 있다. 본 연구에서는 안과 의료진이 질병을 판단할 때 보조적 도움이 될 수 있도록 객관적 진단결과를 제시해주는 각막궤양 의미론적 분할방법에 대하여 제안하였다. 이를 위해 DeepLab 모델을 활용하였고 그 중 Backbone network으로 Xception과 ResNet 네트워크를 이용하였다. 실험결과를 나타내기 위한 평가지표로 다이스 유사계수와 IoU 값을 이용하였고 ResNet101 네트워크를 사용하였을 때 'crop & resized' 이미지에 대해 최대 평균 정확도 93%의 다이스 유사계수 값을 보였다. 본 연구는 객체 검출을 위한 의미론적 분할모델 또한 안구의 각막궤양 부분과 같은 불규칙하고 특이한 모양을 추출하고 분류하는데 뛰어난 결과를 도출할 수 있는 성능을 보유하고 있음을 보여주었다. 향후 학습용 Dataset을 양적으로 보강하여 실험결과의 정확도를 제고할 수 있도록 하고 실제 의료진단 환경에서 구현되어 사용되어 질 수 있도록 할 계획이다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권3호
• /
• pp.1450-1463
• /
• 2019

The problem towards crowd behavior recognition in a serious clustered scene is extremely challenged on account of variable scales with non-uniformity. This paper aims to propose a crowed behavior classification framework based on a transferring hybrid network blending 3D res-net with inception-v3. First, the 3D res-inception network is presented so as to learn the augmented visual feature of UCF 101. Then the target dataset is applied to fine-tune the network parameters in an attempt to classify the behavior of densely crowded scenes. Finally, a transferred entropy function is used to calculate the probability of multiple labels in accordance with these features. Experimental results show that the proposed method could greatly improve the accuracy of crowd behavior recognition and enhance the accuracy of multiple label classification.

• 대한건축학회논문집:구조계
• /
• 제36권3호
• /
• pp.113-120
• /
• 2020

The purpose of this study is to compare the models of Deep Learning-based Convolution Neural Network(CNN) for concrete crack detection. The comparison models are AlexNet, GoogLeNet, VGG16, VGG19, ResNet-18, ResNet-50, ResNet-101, and SqueezeNet which won ImageNet Large Scale Visual Recognition Challenge(ILSVRC). To train, validate and test these models, we constructed 3000 training data and 12000 validation data with 256×256 pixel resolution consisting of cracked and non-cracked images, and constructed 5 test data with 4160×3120 pixel resolution consisting of concrete images with crack. In order to increase the efficiency of the training, transfer learning was performed by taking the weight from the pre-trained network supported by MATLAB. From the trained network, the validation data is classified into crack image and non-crack image, yielding True Positive (TP), True Negative (TN), False Positive (FP), False Negative (FN), and 6 performance indicators, False Negative Rate (FNR), False Positive Rate (FPR), Error Rate, Recall, Precision, Accuracy were calculated. The test image was scanned twice with a sliding window of 256×256 pixel resolution to classify the cracks, resulting in a crack map. From the comparison of the performance indicators and the crack map, it was concluded that VGG16 and VGG19 were the most suitable for detecting concrete cracks.

• 한국멀티미디어학회논문지
• /
• 제24권8호
• /
• pp.1026-1034
• /
• 2021

Tire wear and defect are important factors for safe driving condition. These defects are generally inspected by some specialized experts or very expensive equipments such as stereo depth camera and depth gauge. In this paper, we propose tire safety vision inspector based on deep neural network (DNN). The status of tire wear is categorized into three: 'safety', 'warning', and 'danger' based on depth of tire tread. We propose an attention mechanism for emphasizing the feature of tread area. The attention-based feature is concatenated to output feature maps of the last convolution layer of ResNet-101 to extract more robust feature. Through experiments, the proposed tire wear classification model improves 1.8% of accuracy compared to the existing ResNet-101 model. For detecting the tire defections, the developed tire defect detection model shows up-to 91% of accuracy using the Mask R-CNN model. From these results, we can see that the suggested models are useful for checking on the safety condition of working tire in real environment.

• International Journal of Computer Science & Network Security
• /
• 제22권4호
• /
• pp.101-110
• /
• 2022

Histopathological analysis of biopsy specimens is still used for diagnosis and classifying the brain tumors today. The available procedures are intrusive, time consuming, and inclined to human error. To overcome these disadvantages, need of implementing a fully automated deep learning-based model to classify brain tumor into multiple classes. The proposed CNN model with an accuracy of 92.98 % for categorizing tumors into five classes such as normal tumor, glioma tumor, meningioma tumor, pituitary tumor, and metastatic tumor. Using the grid search optimization approach, all of the critical hyper parameters of suggested CNN framework were instantly assigned. Alex Net, Inception v3, Res Net -50, VGG -16, and Google - Net are all examples of cutting-edge CNN models that are compared to the suggested CNN model. Using huge, publicly available clinical datasets, satisfactory classification results were produced. Physicians and radiologists can use the suggested CNN model to confirm their first screening for brain tumor Multi-classification.

• 대한의용생체공학회:의공학회지
• /
• 제41권2호
• /
• pp.101-106
• /
• 2020

Although benign gastric ulcers do not develop into gastric cancer, they are similar to early gastric cancer and difficult to distinguish. This may lead to misconsider early gastric cancer as gastric ulcer while diagnosing. Since gastric cancer does not have any special symptoms until discovered, it is important to detect gastric ulcers by early gastroscopy to prevent the gastric cancer. Therefore, we developed a Convolution Neural Network (CNN) model that can be helpful for endoscopy. 3,015 images of gastroscopy of patients undergoing endoscopy at Gachon University Gil Hospital were used in this study. Using ResNet-50, three models were developed to classify normal and gastric ulcers, normal and gastric cancer, and gastric ulcer and gastric cancer. We applied the data augmentation technique to increase the number of training data and examined the effect on accuracy by varying the multiples. The accuracy of each model with the highest performance are as follows. The accuracy of normal and gastric ulcer classification model was 95.11% when the data were increased 15 times, the accuracy of normal and gastric cancer classification model was 98.28% when 15 times increased likewise, and 5 times increased data in gastric ulcer and gastric cancer classification model yielded 87.89%. We will collect additional specific shape of gastric ulcer and cancer data and will apply various image processing techniques for visual enhancement. Models that classify normal and lesion, which showed relatively high accuracy, will be re-learned through optimal parameter search.

• 한국안전학회지
• /
• 제34권6호
• /
• pp.50-57
• /
• 2019

In this study, various types of deep learning models that have been proposed recently are classified according to data input / output types and analyzed to find the deep learning model suitable for constructing a crack detection model. First the deep learning models are classified into image classification model, object segmentation model, object detection model, and instance segmentation model. ResNet-101, DeepLab V2, Faster R-CNN, and Mask R-CNN were selected as representative deep learning model of each type. For the comparison, ResNet-101 was implemented for all the types of deep learning model as a backbone network which serves as a main feature extractor. The four types of deep learning models were trained with 500 crack images taken from real concrete structures and collected from the Internet. The four types of deep learning models showed high accuracy above 94% during the training. Comparative evaluation was conducted using 40 images taken from real concrete structures. The performance of each type of deep learning model was measured using precision and recall. In the experimental result, Mask R-CNN, an instance segmentation deep learning model showed the highest precision and recall on crack detection. Qualitative analysis also shows that Mask R-CNN could detect crack shapes most similarly to the real crack shapes.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2019년도 춘계학술발표대회
• /
• pp.455-457
• /
• 2019

In this paper we study the effect of decomposed spatiotemporal convolutions for action recognition in videos. Our motivation emerges from the empirical observation that spatial convolution applied on solo frames of the video provide good performance in action recognition. In this research we empirically show the accuracy of factorized convolution on individual frames of video for action classification. We take 3D ResNet-18 as base line model for our experiment, factorize its 3D convolution to 2D (Spatial) and 1D (Temporal) convolution. We train the model from scratch using Kinetics video dataset. We then fine-tune the model on UCF-101 dataset and evaluate the performance. Our results show good accuracy similar to that of the state of the art algorithms on Kinetics and UCF-101 datasets.

• 경영과정보연구
• /
• 제36권4호
• /
• pp.33-51
• /
• 2017

제4차 산업혁명의 등장과 경제성장으로 인한 '국민 삶의 질 향상' 요구 증대로 인해 의료서비스의 질과 의료비용에 대한 국민들의 요구수준이 향상되고 있으며, 이로 인해 인공지능이 의료현장에 도입되고 있다. 하지만 인공지능이 의료분야에 활용된 사례를 살펴보면 '삶의 질'에 직접적인 영향을 끼치는 만성피부질환에 활용된 사례는 부족한 실정이며, 만성피부질환 중 대표적 질병인 아토피피부염은 정성적 진단 방법으로 인해 진단의 객관성을 확보할 수 없다는 한계가 존재한다. 본 연구에서는 아토피피부염의 객관적 중증도 평가 방법을 마련하여 아토피피부염 환자의 삶의 질을 향상시키고자 다음과 같은 연구를 수행하였다. 첫째, 가톨릭대학교 의과대학 성모병원의 데이터베이스로부터 아토피피부염 환자의 이미지 데이터를 수집했으며, 수집된 이미지 데이터에 대한 정제 및 라벨링 작업을 수행하여 모델 학습과 검증에 적합한 데이터를 확보했다. 둘째, 지능형 아토피피부염 중증도 진단 모형에 적합한 이미지 인식 알고리즘을 파악하기 위해 다양한 CNN 알고리즘들을 병변별 학습용 데이터로 학습시키고, 검증용 데이터를 활용하여 해당 모델의 이미지 인식 정확도를 측정했다. 실증분석 결과 홍반(Erythema)의 경우 'ResNet V1 101', 긁은 정도(Excoriation)의 경우 'ResNet V2 50'이 90% 이상의 정확도를 기록하였으며, 태선화(Lichenification)의 경우 학습용 데이터 부족의 한계로 인해 두 병변보다 낮은 89%의 정확도를 보였다. 해당 결과를 통해 이미지 인식 알고리즘이 단순한 사물 인식 분야뿐만 아니라 전문적 지식이 요구되는 분야에도 높은 성능을 나타낸다는 것을 실증적으로 입증했으며, 본 연구는 실제 아토피피부염 환자의 이미지 데이터를 활용했다는 측면에서 실제 임상환경에서 활용성이 높을 것으로 사료된다.