• 스마트미디어저널
• /
• 제10권2호
• /
• pp.22-29
• /
• 2021

In the past decade, deep learning has been applied to various medical image analysis tasks. Skeletal bone age estimation is clinically important as it can help prevent age-related illness and pave the way for new anti-aging therapies. Recent research has applied deep learning techniques to the task of bone age assessment and achieved positive results. In this paper, we propose a bone age prediction method using a deep convolutional neural network. Specifically, we first train a classification model that automatically localizes the most discriminative region of an image and crops it from the original image. The regions of interest are then used as input for a regression model to estimate the age of the patient. The experiments are conducted on a whole-body scintigraphy dataset that was collected by Chonnam National University Hwasun Hospital. The experimental results illustrate the potential of our proposed method, which has a mean absolute error of 3.35 years. Our proposed framework can be used as a robust supporting tool for clinicians to prevent age-related diseases.

• 농업과학연구
• /
• 제49권4호
• /
• pp.785-793
• /
• 2022

In this study, tomato maturity was estimated based on deep learning for a harvesting robot. Tomato images were obtained using a RGB camera installed on a monitoring robot, which was developed previously, and the samples were cropped to 128 × 128 size images to generate a dataset for training the classification model. The classification model was constructed based on convolutional neural networks, and the mean-variance loss was used to learn implicitly the distribution of the data features by class. In the test stage, the tomato maturity was estimated as a continuous index, which has a range of 0 to 1, by calculating the expected class value. The results show that the F1-score of the classification was approximately 0.94, and the performance was similar to that of a deep learning-based classification task in the agriculture field. In addition, it was possible to estimate the distribution in each maturity stage. From the results, it was found that our approach can not only classify the discrete maturation stages of the tomatoes but also can estimate the continuous maturity.

• 한국지능시스템학회논문지
• /
• 제26권2호
• /
• pp.93-98
• /
• 2016

정확한 강수예측을 위해서는 예측인자 선정과 예측방법에 대한 선택이 매우 중요하다. 최근에는 강수예측 방법으로 기계학습 기법이 많이 사용되고 있으며, 그 중에서도 특히 인공신경망을 사용한 강수예측 방법은 좋은 성능을 보였다. 본 논문에서는 딥러닝 기법 중 하나인 DBN(deep belief network)를 이용한 새로운 강수예측 방법을 제안한다. DBN는 비지도 사전 학습을 통해 초기 가중치를 설정하여 기존 인공신경망의 문제점을 보완한다. 예측인자로는 기온, 전일-전주 강수일, 태양과 달 궤도 관련 자료를 선정하였다. 기온과 전일-전주 강수일은 서울에서의 1974년부터 2013년까지 총 40년간의 AWS(automatic weather system) 관측 자료를 사용하였고, 태양과 달의 궤도 관련 자료는 서울을 중심으로 계산한 결과를 사용하였다. 전체 기간에서 일부는 학습 자료로 사용하여 예측모델을 생성하였고, 나머지를 생성한 모델의 검증 자료로 사용하였다. 모델 검증 결과로 나온 예측값들은 확률값을 가지며 임계치를 이용하여 강수유무를 판별하였다. 강수 정확도의 척도로 양분예보기법 중 CSI(critical successive index)와 Bias(frequency bias)를 계산하였다. 이를 통해 DBN와 MLP(multilayer perceptron)의 성능을 비교한 결과 DBN의 강수 예측 정확도가 높았고, 수행속도 또한 2배 이상 빨랐다.

• 문화기술의 융합
• /
• 제9권4호
• /
• pp.487-494
• /
• 2023

본 연구에서는 무료 딥러닝 도구인 R과 텐서플로우에 대한 성능 비교를 수행하였다. 실험에서는 각 도구를 사용하여 6종류의 심층 신경망을 구축하고 10년간의 한국 온도 데이터셋을 사용하여 신경망을 학습시켰다. 구축된 신경망의 입력층 노드 갯수는 10개, 출력층은 5개로 설정 하였으며, 은닉층은 5, 10, 20개로 설정하여 실험을 진행 하였다. 학습 데이터는 2013년 3월 1일부터 2023년 3월 29일까지 서울시 강남구에서 수집된 온도 데이터 3681건을 사용하였다. 성능 비교를 위해, 학습된 신경망을 사용하여, 5일간의 온도를 예측하고 예측된 값과 실제값을 사용하여 평균 제곱근 오차(root mean square error, RMSE)값을 측정하였다. 실험결과, 은닉층이 1개인 경우, R의 학습 오차는 0.04731176이었으며, 텐서플로우는 0.06677193으로 측정되었으며, 은닉층이 2개인 경우에는 R이 0.04782134, 텐서플로 우는 0.05799060로 측정되었다. 전체적으로 R이 더 우수한 성능을 보였다. 우리는 기계학습을 처음 접하는 사용자들에게 두 도구에 대한 정량적 성능 정보를 제공함으로써, 도구 선택에서 발생하는 어려움을 해소하고자 하였다.

• 한국멀티미디어학회논문지
• /
• 제25권8호
• /
• pp.1212-1223
• /
• 2022

Skin lesions are common diseases that range from skin rashes to skin cancer, which can lead to death. Note that early diagnosis of skin diseases can be important because early diagnosis of skin diseases considerably can reduce the course of treatment and the harmful effect of the disease. Recently, the development of computer-aided diagnosis (CAD) systems based on artificial intelligence has been actively made for the early diagnosis of skin diseases. In a typical CAD system, the accurate classification of skin lesion types is of great importance for improving the diagnosis performance. Motivated by this, we propose a novel deep ensemble classification with multi-scale attention networks. The proposed deep ensemble networks are jointly trained using a single loss function in an end-to-end manner. In addition, the proposed deep ensemble network is equipped with a multi-scale attention mechanism and segmentation information of the original skin input image, which improves the classification performance. To demonstrate our method, the publicly available human skin disease dataset (HAM 10000) and the private animal skin lesion dataset were used for the evaluation. Experiment results showed that the proposed methods can achieve 97.8% and 81% accuracy on each HAM10000 and animal skin lesion dataset. This research work would be useful for developing a more reliable CAD system which helps doctors early diagnose skin diseases.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2018년도 추계학술대회
• /
• pp.201-202
• /
• 2018

본 논문에서는 기존 ImageNet과 Oxford-IIIT Pet Image의 Dataset에서 제공하는 개 품종 이미지와 인터넷 상에서 개 품종 이미지를 데이터 마이닝을 통해 획득된 개 품종 이미지를 결합하고 Random-Label을 추가 하여 개 품종 122개의 클래스와 개 품종이 아닌 1개의 클래스를 인식하는 방법에 대해 소개 한다. 기존 DB만을 사용하였을 때 개 품종 인식률 대비 기존 DB와 수집 DB를 모두 사용한 개 품종 인식률이 Top-1에 대해서 1.5% 개선되었다. 개가 아닌 이미지 인식은 랜덤 DB를 10000장의 경우 93% 인식률을 확인했다.

• International Journal of Contents
• /
• 제18권2호
• /
• pp.68-80
• /
• 2022

The health of the human heart is commonly measured using ECG (Electrocardiography) signals. To identify any anomaly in the human heart, the time-sequence of ECG signals is examined manually by a cardiologist or cardiac electrophysiologist. Lightweight anomaly detection on ECG signals in an embedded system is expected to be popular in the near future, because of the increasing number of heart disease symptoms. Some previous research uses deep learning networks such as LSTM and BiLSTM to detect anomaly signals without any handcrafted feature. Unfortunately, lightweight LSTMs show low precision and heavy LSTMs require heavy computing powers and volumes of labeled dataset for symptom classification. This paper proposes an ECG anomaly detection system based on two level BiLSTM for acceptable precision with lightweight networks, which is lightweight and usable at home. Also, this paper presents a new threshold technique which considers statistics of the current ECG pattern. This paper's proposed model with BiLSTM detects ECG signal anomaly in 0.467 ~ 1.0 F₁ score, compared to 0.426 ~ 0.978 F₁ score of the similar model with LSTM except one highly noisy dataset.

• 대한원격탐사학회지
• /
• 제39권4호
• /
• pp.395-407
• /
• 2023

Super-resolution (SR) has great significance in image processing because it enables downstream vision tasks with high spatial resolution. Recently, SR studies have adopted deep learning networks and achieved remarkable SR performance compared to conventional example-based methods. Deep-learning-based SR models generally require low-resolution (LR) images and the corresponding high-resolution (HR) images as training dataset. Due to the difficulties in obtaining real-world LR-HR datasets, most SR models have used only HR images and generated LR images with predefined degradation such as bicubic downsampling. However, SR models trained on simple image degradation do not reflect the properties of the images and often result in deteriorated SR qualities when applied to real-world images. In this study, we propose an image degradation model for HR satellite images based on the modulation transfer function (MTF) of an imaging sensor. Because the proposed method determines the image degradation based on the sensor properties, it is more suitable for training SR models on remote sensing images. Experimental results on HR satellite image datasets demonstrated the effectiveness of applying MTF-based filters to construct a more realistic LR-HR training dataset.

• ETRI Journal
• /
• 제46권2호
• /
• pp.205-217
• /
• 2024

We propose a bandwidth prediction approach based on deep learning. The approach is intended to accurately predict the bandwidth of various types of mobile networks. We first use a machine learning technique, namely, the gradient boosting algorithm, to recognize the connected mobile network. Second, we apply a handover detection algorithm based on network recognition to account for vertical handover that causes the bandwidth variance. Third, as the communication performance offered by 3G, 4G, and 5G networks varies, we suggest a bidirectional long short-term memory model with time distribution for bandwidth prediction per network. To increase the prediction accuracy, pretraining and fine-tuning are applied for each type of network. We use a dataset collected at University College Cork for network recognition, handover detection, and bandwidth prediction. The performance evaluation indicates that the handover detection algorithm achieves 88.5% accuracy, and the bandwidth prediction model achieves a high accuracy, with a root-mean-square error of only 2.12%.

• International Journal of Internet, Broadcasting and Communication
• /
• 제16권1호
• /
• pp.124-131
• /
• 2024

Breast cancer classification using ultrasound, while widely employed, faces challenges due to its relatively low predictive value arising from significant overlap in characteristics between benign and malignant lesions, as well as operator-dependency. To alleviate these challenges and reduce dependency on radiologist interpretation, the implementation of automatic breast cancer classification in ultrasound image can be helpful. To deal with this problem, we propose a semi-supervised deep learning framework for breast cancer classification. In the proposed method, we could achieve reasonable performance utilizing less than 50% of the training data for supervised learning in comparison to when we utilized a 100% labeled dataset for training. Though it requires more modification, this methodology may be able to alleviate the time-consuming annotation burden on radiologists by reducing the number of annotation, contributing to a more efficient and effective breast cancer detection process in ultrasound images.