• International Journal of Contents
• /
• 제16권2호
• /
• pp.8-18
• /
• 2020

Data is critical in deep learning but the scarcity of data often occurs in research, especially in the preparation of the paired training data. In this paper, document image binarization with unpaired data is studied by introducing adversarial learning, excluding the need for supervised or labeled datasets. However, the simple extension of the previous unpaired training to binarization inevitably leads to poor performance compared to paired data training. Thus, a new deep learning approach is proposed by introducing a multi-diversity of higher quality generated images. In this paper, a two-stage model is proposed that comprises the generative adversarial network (GAN) followed by the U-net network. In the first stage, the GAN uses the unpaired image data to create paired image data. With the second stage, the generated paired image data are passed through the U-net network for binarization. Thus, the trained U-net becomes the binarization model during the testing. The proposed model has been evaluated over the publicly available DIBCO dataset and it outperforms other techniques on unpaired training data. The paper shows the potential of using unpaired data for binarization, for the first time in the literature, which can be further improved to replace paired data training for binarization in the future.

• 재활간호학회지
• /
• 제6권1호
• /
• pp.70-78
• /
• 2003

This study was a quasi-experimental study of nonequivalent control group pretest-posttest design to investigate the effect of rhythmic dance movement training on the physical and psychological functions of the elderly. The data were collected from November, 2001 to February, 2002. The subjects for this study were 34 elderly who was over 65 years old and was living in J city. The elderly selected for this study were: free from heart and pulmonary disease and not regular exercise. The rhythmic dance movement training in watching video tape was rhythmic dance movement and education and supportive care. The rhythmic dance movement was 40-60 intensity, 8 weeks' period, three times a week, 60 minutes a day. The data were analysed by $X^2$-test, paired t-test and unpaired t-test and ANCOVA through SAS/PC program. The results of the study were as follows: 1. There was insignificant difference in the gait length experimental and control group. 2. There was significant difference in the gait speed between the two groups. 3. There was significant difference in the dynamic valance between the two groups. 4. There was no significant difference in the depression between the two groups. 5. There was no significant difference in the Quality of life between the two groups. As shown above, the results of the 8 weeks' rhythmic movement program for the elderly produced positive effects on gait speed, dynamic valance. And this program was expected that it was more effective in different intervention period, verified program. Also it was needed follow study.

• Asian Pacific Journal of Cancer Prevention
• /
• 제16권18호
• /
• pp.8127-8133
• /
• 2016

Background: The objective of this study was to evaluate the effect of back massage on the anxiety state, cortisol level, systolic/diastolic blood pressure, pulse rate, and sleep quality in family caregivers of patients with cancer. Materials and Methods: Forty-four family caregivers were randomly assigned to either the experimental or control group (22 interventions, 22 controls) after they were matched on age and gender. The intervention consisted of back massage for 15 minutes per day for a week. Main research outcomes were measured at baseline (day I) and follow-up (day 7). Unpaired t-test, paired t test and chi-square test were used to analyse data. Results: The majority of the caregivers were women, married, secondary school educated and housewife. State anxiety (p<0.001), cortisol level (p<0.05), systolic/diastolic blood pressure (p<0.001, p<0.01 respectively), and pulse rate (p<0.01) were significantly decreased, and sleep quality (p<0.001) increased after back massage intervention. Conclusions: The study results show that family caregivers for patients with cancer can benefit from back massage to improve state anxiety, cortisol level, blood pressure and heart rate, and sleep quality. Oncology nurses can take advantage of back massage, which is non-pharmacologic and easily implemented method, as an independent nursing action to support caregivers for patients with cancer.

• 대한의용생체공학회:의공학회지
• /
• 제45권3호
• /
• pp.118-127
• /
• 2024

Commonly deep learning methods for enhancing the quality of medical images use unpaired dataset due to the impracticality of acquiring paired dataset through commercial imaging system. In this paper, we propose a supervised learning method to enhance the quality of ultrasound images. The U-net model is designed by incorporating a divide-and-conquer approach that divides and processes an image into four parts to overcome data shortage and shorten the learning time. The proposed model is trained using paired dataset consisting of 828 pairs of low-quality and high-quality images with a resolution of 512x512 pixels obtained by varying the number of channels for the same subject. Out of a total of 828 pairs of images, 684 pairs are used as the training dataset, while the remaining 144 pairs served as the test dataset. In the test results, the average Mean Squared Error (MSE) was reduced from 87.6884 in the low-quality images to 45.5108 in the restored images. Additionally, the average Peak Signal-to-Noise Ratio (PSNR) was improved from 28.7550 to 31.8063, and the average Structural Similarity Index (SSIM) was increased from 0.4755 to 0.8511, demonstrating significant enhancements in image quality.

• 한국멀티미디어학회논문지
• /
• 제25권1호
• /
• pp.1-9
• /
• 2022

Ship detection in synthetic aperture radar (SAR) images is an important application in marine monitoring for the military and civilian domains. Over the past decade, object detection has achieved significant progress with the development of convolutional neural networks (CNNs) and lot of labeled databases. However, due to difficulty in collecting and labeling SAR images, it is still a challenging task to solve SAR ship detection CNNs. To overcome the problem, some methods have employed conventional data augmentation techniques such as flipping, cropping, and affine transformation, but it is insufficient to achieve robust performance to handle a wide variety of types of ships. In this paper, we present a novel and effective approach for deep SAR ship detection, that exploits label-rich Electro-Optical (EO) images. The proposed method consists of two components: a data augmentation network and a ship detection network. First, we train the data augmentation network based on conditional generative adversarial network (cGAN), which aims to generate additional SAR images from EO images. Since it is trained using unpaired EO and SAR images, we impose the cycle-consistency loss to preserve the structural information while translating the characteristics of the images. After training the data augmentation network, we leverage the augmented dataset constituted with real and translated SAR images to train the ship detection network. The experimental results include qualitative evaluation of the translated SAR images and the comparison of detection performance of the networks, trained with non-augmented and augmented dataset, which demonstrates the effectiveness of the proposed framework.