• Smart Media Journal
• /
• v.10 no.4
• /
• pp.45-54
• /
• 2021

The skin is the body's first line of defense against external infection. When a skin disease strikes, the skin's protective role is compromised, necessitating quick diagnosis and treatment. Recently, as artificial intelligence has advanced, research for technical applications has been done in a variety of sectors, including dermatology, to reduce the rate of misdiagnosis and obtain quick treatment using artificial intelligence. Although previous studies have diagnosed skin diseases with low incidence, this paper proposes a method to classify common illnesses such as warts and corns using a convolutional neural network. The data set used consists of 3 classes and 2,515 images, but there is a problem of lack of training data and class imbalance. We analyzed the performance using a deep metric loss function and a cross-entropy loss function to train the model. When comparing that in terms of accuracy, recall, F1 score, and accuracy, the former performed better.

• The Journal of the Acoustical Society of Korea
• /
• v.40 no.2
• /
• pp.176-182
• /
• 2021

This paper evaluates and compares the performance of the Deep Nerual Network (DNN)-based speech enhancement models according to various loss functions. We used a complex network that can consider the phase information of speech as a baseline model. As the loss function, we consider two types of basic loss functions; the Mean Squared Error (MSE) and the Scale-Invariant Source-to-Noise Ratio (SI-SNR), and two types of perceptual-based loss functions, including the Perceptual Metric for Speech Quality Evaluation (PMSQE) and the Log Mel Spectra (LMS). The performance comparison was performed through objective evaluation and listening tests with outputs obtained using various combinations of the loss functions. Test results show that when a perceptual-based loss function was combined with MSE or SI-SNR, the overall performance is improved, and the perceptual-based loss functions, even exhibiting lower objective scores showed better performance in the listening test.

• Journal of the Korean Society of Radiology
• /
• v.16 no.5
• /
• pp.595-602
• /
• 2022

This study tried to suggest the most efficient learning rate for accurate and efficient automatic diagnosis of medical images for chest X-ray pneumonia images using deep learning. After setting the learning rates to 0.1, 0.01, 0.001, and 0.0001 in the Inception V3 deep learning model, respectively, deep learning modeling was performed three times. And the average accuracy and loss function value of verification modeling, and the metric of test modeling were set as performance evaluation indicators, and the performance was compared and evaluated with the average value of three times of the results obtained as a result of performing deep learning modeling. As a result of performance evaluation for deep learning verification modeling performance evaluation and test modeling metric, modeling with a learning rate of 0.001 showed the highest accuracy and excellent performance. For this reason, in this paper, it is recommended to apply a learning rate of 0.001 when classifying the presence or absence of pneumonia on chest X-ray images using a deep learning model. In addition, it was judged that when deep learning modeling through the application of the learning rate presented in this paper could play an auxiliary role in the classification of the presence or absence of pneumonia on chest X-ray images. In the future, if the study of classification for diagnosis and classification of pneumonia using deep learning continues, the contents of this thesis research can be used as basic data, and furthermore, it is expected that it will be helpful in selecting an efficient learning rate in classifying medical images using artificial intelligence.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.49-54
• /
• 2024

Data imbalance is a common issue encountered in classification problems, stemming from a significant disparity in the number of samples between classes within the dataset. Such data imbalance typically leads to problems in classification models, including overfitting, underfitting, and misinterpretation of performance metrics. Methods to address this issue include resampling, augmentation, regularization techniques, and adjustment of loss functions. In this paper, we focus on loss function adjustment, particularly comparing the performance of various configurations of loss functions (Cross Entropy, Balanced Cross Entropy, two settings of Focal Loss: 𝛼 = 1 and 𝛼 = Balanced, Asymmetric Loss) on Multi-Class black-box video data with imbalance issues. The comparison is conducted using the I3D, and R3D_18 models.

• Journal of the Korean Society of Radiology
• /
• v.15 no.4
• /
• pp.547-554
• /
• 2021

In order to quickly and accurately diagnose pneumonia on a chest X-ray image, different batch sizes of 4, 8, 16, and 32 were applied to the same Xception deep learning model, and modeling was performed 3 times, respectively. As a result of the performance evaluation of deep learning modeling, in the case of modeling to which batch size 32 was applied, the results of accuracy, loss function value, mean square error, and learning time per epoch showed the best results. And in the accuracy evaluation of the Test Metric, the modeling applied with batch size 8 showed the best results, and the precision evaluation showed excellent results in all batch sizes. In the recall evaluation, modeling applied with batch size 16 showed the best results, and for F1-score, modeling applied with batch size 16 showed the best results. And the AUC score evaluation was the same for all batch sizes. Based on these results, deep learning modeling with batch size 32 showed high accuracy, stable artificial neural network learning, and excellent speed. It is thought that accurate and rapid lesion detection will be possible if a batch size of 32 is applied in an automatic diagnosis study for feature extraction and classification of pneumonia in chest X-ray images using deep learning in the future.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.3
• /
• pp.881-895
• /
• 2023

Modern image inpainting techniques based on deep learning have achieved remarkable performance, and more and more people are working on repairing more complex and larger missing areas, although this is still challenging, especially for facial image inpainting. For a face image with a huge missing area, there are very few valid pixels available; however, people have an ability to imagine the complete picture in their mind according to their subjective will. It is important to simulate this capability while maintaining the identity features of the face as much as possible. To achieve this goal, we propose a three-stage network model, which we refer to as the identity and structure feature refinement network (ISFRNet). ISFRNet is based on 1) a pre-trained pSp-styleGAN model that generates an extremely realistic face image with rich structural features; 2) a shallow structured network with a small receptive field; and 3) a modified U-net with two encoders and a decoder, which has a large receptive field. We choose structural similarity index (SSIM), peak signal-to-noise ratio (PSNR), L1 Loss and learned perceptual image patch similarity (LPIPS) to evaluate our model. When the missing region is 20%-40%, the above four metric scores of our model are 28.12, 0.942, 0.015 and 0.090, respectively. When the lost area is between 40% and 60%, the metric scores are 23.31, 0.840, 0.053 and 0.177, respectively. Our inpainting network not only guarantees excellent face identity feature recovery but also exhibits state-of-the-art performance compared to other multi-stage refinement models.

• Journal of Korea Multimedia Society
• /
• v.24 no.7
• /
• pp.880-889
• /
• 2021

In everyday life, recognizing people's emotions from their frames is essential and is a popular research domain in the area of computer vision. Visual emotion has a severe class imbalance in which most of the data are distributed in specific categories. The existing methods do not consider class imbalance and used accuracy as the performance metric, which is not suitable for evaluating the performance of the imbalanced dataset. Therefore, we proposed a method for recognizing visual emotion using balanced data augmentation to address the class imbalance. The proposed method generates a balanced dataset by adopting the random over-sampling and image transformation methods. Also, the proposed method uses the Focal loss as a loss function, which can mitigate the class imbalance by down weighting the well-classified samples. EfficientNet, which is the state-of-the-art method for image classification is used to recognize visual emotion. We compare the performance of the proposed method with that of conventional methods by using a public dataset. The experimental results show that the proposed method increases the F1 score by 40% compared with the method without data augmentation, mitigating class imbalance without loss of classification accuracy.

• Journal of Broadcast Engineering
• /
• v.26 no.7
• /
• pp.868-876
• /
• 2021

Deploying deep neural network model training performs remarkable performance in the fields of Object detection and Instance segmentation. To train these models, features are first extracted from the input image using a backbone network. The extracted features can be reused by various tasks. Research has been actively conducted to serve various tasks by using these learned features. In this process, standardization discussions about encoding, decoding, and transmission methods are proceeding actively. In this scenario, it is necessary to analyze the response characteristics of features against various distortions that may occur in the data transmission or data compression process. In this paper, experiment was conducted to inject various distortions into the feature in the object recognition task. And analyze the mAP (mean Average Precision) metric between the predicted value output from the neural network and the target value as the intensity of various distortions was increased. Experiments have shown that features are more robust to distortion than images. And this points out that using the feature as transmission means can prevent the loss of information against the various distortions during data transmission and compression process.

• Korean Journal of Remote Sensing
• /
• v.39 no.1
• /
• pp.1-21
• /
• 2023

In this technically advanced era, we are surrounded by smartphones, computers, and cameras, which help us to store visual information in 2D image planes. However, such images lack 3D spatial information about the scene, which is very useful for scientists, surveyors, engineers, and even robots. To tackle such problems, depth maps are generated for respective image planes. Depth maps or depth images are single image metric which carries the information in three-dimensional axes, i.e., xyz coordinates, where z is the object's distance from camera axes. For many applications, including augmented reality, object tracking, segmentation, scene reconstruction, distance measurement, autonomous navigation, and autonomous driving, depth estimation is a fundamental task. Much of the work has been done to calculate depth maps. We reviewed the status of depth map estimation using different techniques from several papers, study areas, and models applied over the last 20 years. We surveyed different depth-mapping techniques based on traditional ways and newly developed deep-learning methods. The primary purpose of this study is to present a detailed review of the state-of-the-art traditional depth mapping techniques and recent deep learning methodologies. This study encompasses the critical points of each method from different perspectives, like datasets, procedures performed, types of algorithms, loss functions, and well-known evaluation metrics. Similarly, this paper also discusses the subdomains in each method, like supervised, unsupervised, and semi-supervised methods. We also elaborate on the challenges of different methods. At the conclusion of this study, we discussed new ideas for future research and studies in depth map research.

• Korean Journal of Remote Sensing
• /
• v.39 no.2
• /
• pp.143-156
• /
• 2023

To precisely and reliably analyze the contents of the satellite imagery, recognizing the clouds which are the obstacle to gathering the useful information is essential. In recent times, deep learning yielded satisfactory results in various tasks, so many studies using deep neural networks have been conducted to improve the performance of cloud detection. However, existing methods for cloud detection have the limitation on increasing the performance due to the adopting the network models for semantic image segmentation without modification. To tackle this problem, we introduced the multi-branch neural architecture search to find optimal network structure for cloud detection. Additionally, the proposed method adopts the soft intersection over union (IoU) as loss function to mitigate the disagreement between the loss function and the evaluation metric and uses the various data augmentation methods. The experiments are conducted using the cloud detection dataset acquired by Arirang-3/3A satellite imagery. The experimental results showed that the proposed network which are searched network architecture using cloud dataset is 4% higher than the existing network model which are searched network structure using urban street scenes with regard to the IoU. Also, the experimental results showed that the soft IoU exhibits the best performance on cloud detection among the various loss functions. When comparing the proposed method with the state-of-the-art (SOTA) models in the field of semantic segmentation, the proposed method showed better performance than the SOTA models with regard to the mean IoU and overall accuracy.