• The Journal of Korea Robotics Society
• /
• v.15 no.2
• /
• pp.184-189
• /
• 2020

This paper is a study on data augmentation for small dataset by using deep learning. In case of training a deep learning model for recognition and classification of non-mainstream objects, there is a limit to obtaining a large amount of training data. Therefore, this paper proposes a data augmentation method using perspective transform and image synthesis. In addition, it is necessary to save the object area for all training data to detect the object area. Thus, we devised a way to augment the data and save object regions at the same time. To verify the performance of the augmented data using the proposed method, an experiment was conducted to compare classification accuracy with the augmented data by the traditional method, and transfer learning was used in model learning. As experimental results, the model trained using the proposed method showed higher accuracy than the model trained using the traditional method.

• The Journal of the Convergence on Culture Technology
• /
• v.9 no.2
• /
• pp.561-568
• /
• 2023

Various defects occur in concrete structures due to internal and external environments. If there is a defect, it is important to efficiently identify and maintain it because there is a problem with the structural safety of concrete. However, recent deep learning research has focused on cracks in concrete, and studies on exfoliation and contamination are lacking. In this study, focusing on exfoliation and contamination, which are difficult to label, four models were developed and their performance evaluated through unlabelling method, filtering method, the convergence of transfer learning based k-means clustering. As a result of the analysis, the convergence model classified the defects in the most detail and could increase the efficiency compared to direct labeling. It is hoped that the results of this study will contribute to the development of deep learning models for various types of defects that are difficult to label in the future.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.8
• /
• pp.31-38
• /
• 2023

Style transfer is one of deep learning-based image processing techniques that has been actively researched recently. These research efforts have led to significant improvements in the quality of result images. Style transfer is a technology that takes a content image and a style image as inputs and generates a transformed result image by applying the characteristics of the style image to the content image. It is becoming increasingly important in exploiting the diversity of digital content. To improve the usability of style transfer technology, ensuring stable performance is crucial. Recently, in the field of natural language processing, the concept of Transformers has been actively utilized. Attention maps, which forms the basis of Transformers, is also being actively applied and researched in the development of style transfer techniques. In this paper, we analyze the representative techniques SANet and AdaAttN and propose a novel attention map-based structure which can generate improved style transfer results. The results demonstrate that the proposed technique effectively preserves the structure of the content image while applying the characteristics of the style image.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.7
• /
• pp.3511-3532
• /
• 2019

Coverage-guided fuzzing is an efficient solution that has been widely used in software testing. By guiding fuzzers through the coverage information, seeds that generate new paths will be retained to continually increase the coverage. However, we observed that most samples follow the same few high-frequency paths. The seeds that exercise a high-frequency path are saved for the subsequent mutation process until the user terminates the test process, which directly affects the efficiency with which the low-frequency paths are tested. In this paper, we propose a fuzzing solution, ER-Fuzz, that truncates the recording of a high-frequency path to influence coverage. It utilizes a deep learning-based classifier to locate the high and low-frequency path transfer points; then, it instruments at the transfer position to promote the probability low-frequency transfer paths while eliminating subsequent variations of the high-frequency path seeds. We implemented a prototype of ER-Fuzz based on the popular fuzzer AFL and evaluated it on several applications. The experimental results show that ER-Fuzz improves the coverage of the original AFL method to different degrees. In terms of the number of crash discoveries, in the best case, ER-Fuzz found 115% more unique crashes than did AFL. In total, seven new bugs were found and new CVEs were assigned.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.8
• /
• pp.81-88
• /
• 2017

This paper proposes a method of detecting rain regions from a single image. More specifically, a way of training the deep convolutional neural network based on the collected rain and non-rain patches is presented in a supervised manner. It is also shown that the proposed rain detection method based on deep convolutional neural network can provide better performance than the conventional rain detection method based on dictionary learning. Moreover, it is confirmed that the application of the proposed rain detection for rain removal can lead to some improvement in detail representation on the low-frequency regions of the rain-removed images. Additionally, this paper introduces the rain transfer method that inserts rain patterns into original images, thereby producing rain effects on the resulting images. The proposed rain transfer method could be used to augment rain patterns while constructing rain database.

• International Journal of Computer Science & Network Security
• /
• v.22 no.5
• /
• pp.175-181
• /
• 2022

COVID-19 is an acute respiratory syndrome that affects the host's breathing and respiratory system. The novel disease's first case was reported in 2019 and has created a state of emergency in the whole world and declared a global pandemic within months after the first case. The disease created elements of socioeconomic crisis globally. The emergency has made it imperative for professionals to take the necessary measures to make early diagnoses of the disease. The conventional diagnosis for COVID-19 is through Polymerase Chain Reaction (PCR) testing. However, in a lot of rural societies, these tests are not available or take a lot of time to provide results. Hence, we propose a COVID-19 classification system by means of machine learning and transfer learning models. The proposed approach identifies individuals with COVID-19 and distinguishes them from those who are healthy with the help of Deep Visual Embeddings (DVE). Five state-of-the-art models: VGG-19, ResNet50, Inceptionv3, MobileNetv3, and EfficientNetB7, were used in this study along with five different pooling schemes to perform deep feature extraction. In addition, the features are normalized using standard scaling, and 4-fold cross-validation is used to validate the performance over multiple versions of the validation data. The best results of 88.86% UAR, 88.27% Specificity, 89.44% Sensitivity, 88.62% Accuracy, 89.06% Precision, and 87.52% F1-score were obtained using ResNet-50 with Average Pooling and Logistic regression with class weight as the classifier.

• International Journal of Advanced Culture Technology
• /
• v.10 no.1
• /
• pp.302-309
• /
• 2022

Occupational safety accidents are caused by various factors, and it is difficult to predict when and why they occur, and it is directly related to the lives of workers, so the interest in safety accidents is increasing every year. Therefore, in order to reduce safety accidents at industrial fields, workers are required to wear personal protective equipment. In this paper, we proposes a method to automatically check whether workers are wearing safety helmets among the protective equipment in the industrial field. It detects whether or not the helmet is worn using YOLOv5, a computer vision-based deep learning object detection algorithm. We transfer learning the s model among Yolov5 models with different learning rates and epochs, evaluate the performance, and select the optimal model. The selected model showed a performance of 0.959 mAP.

• KIISE Transactions on Computing Practices
• /
• v.24 no.1
• /
• pp.53-57
• /
• 2018

A poisoning accident is often caused by a situation in which people eat poisonous mushrooms because they cannot distinguish between edible mushrooms and poisonous mushrooms. In this paper, we propose an automatic mushroom recognition system by using the convolutional neural network. We collected 1478 mushroom images of 38 species using image crawling, and used the dataset for learning the convolutional neural network. A comparison experiment using AlexNet, VGGNet, and GoogLeNet was performed using the collected datasets, and a comparison experiment using a class number expansion and a fine-tuning technique for transfer learning were performed. As a result of our experiment, we achieve 82.63% top-1 accuracy and 96.84% top-5 accuracy on test set of our dataset.

• Journal of Electrical Engineering and Technology
• /
• v.13 no.5
• /
• pp.2107-2115
• /
• 2018

Over recent years, high-concentration of particulate matters (e.g., a.k.a. fine dust) in South Korea has increasingly evoked considerable concerns about public health. It is intractable to track and report $PM_{10}$ measurements to the public on a real-time basis. Even worse, such records merely amount to averaged particulate concentration at particular regions. Under this circumstance, people are prone to being at risk at rapidly dispersing air pollution. To address this challenge, we attempt to build a predictive model via deep learning to the concentration of particulates ($PM_{10}$). The proposed method learns a binary decision rule on the basis of video sequences to predict whether the level of particulates ($PM_{10}$) in real time is harmful (>$80{\mu}g/m^3$) or not. To our best knowledge, no vision-based $PM_{10}$ measurement method has been proposed in atmosphere research. In experimental studies, the proposed model is found to outperform other existing algorithms in virtue of convolutional deep learning networks. In this regard, we suppose this vision based-predictive model has lucrative potentials to handle with upcoming challenges related to particulate measurement.

• Journal of Biomedical Engineering Research
• /
• v.43 no.2
• /
• pp.102-108
• /
• 2022

Purpose: To propose a deep learning-based clinical decision support technique for laryngeal disease on epiglottis, tongue and vocal cords. Materials and Methods: A total of 873 laryngeal endoscopic images were acquired from the PACS database of Pusan N ational University Yangsan Hospital. and VGG16 model was applied with transfer learning and fine-tuning. Results: The values of precision, recall, accuracy and F1-score for test dataset were 0.94, 0.97, 0.95 and 0.95 for epiglottis images, 0.91, 1.00, 0.95 and 0.95 for tongue images, and 0.90, 0.64, 0.73 and 0.75 for vocal cord images, respectively. Conclusion: Experimental results demonstrated that the proposed model have a potential as a tool for decision-supporting of otolaryngologist during manual inspection of laryngeal endoscopic images.