• Journal of Broadcast Engineering
• /
• v.25 no.6
• /
• pp.854-860
• /
• 2020

Deep neural networks have shown remarkable performance in various areas, including image classification and speech recognition. The variety of data generated by augmentation plays an important role in improving the performance of the neural network. The transformation of data in the augmentation process makes it possible for neural networks to be learned more generally through more diverse forms. In the traditional field of image process, not only new augmentation methods have been proposed for improving the performance, but also exploring methods for an optimal augmentation policy that can be changed according to the dataset and structure of networks. Inspired by the prior work, this paper aims to explore to search for an optimal augmentation policy in the field of sound data. We carried out many experiments randomly combining various augmentation methods such as adding noise, pitch shift, or time stretch to empirically search which combination is most effective. As a result, by applying the optimal data augmentation policy we achieve the improved classification accuracy on the environmental sound classification dataset (ESC-50).

• Archives of Plastic Surgery
• /
• v.42 no.1
• /
• pp.4-10
• /
• 2015

Mammary implants marketed by Poly Implant Proth$\grave{e}$se (PIP) were found to contain industrial grade silicone and this caused heightened anxiety and extensive publicity regarding their safety in humans. These implants were used in a large number of patients worldwide for augmentation or breast reconstruction. We reviewed articles identified by searches of Medline, PubMed, Embase, and Google Scholar databases up to May 2014 using the terms: "PIP", "Poly Implant Proth$\grave{e}$se", "breast implants" and "augmentation mammoplasty" "siloxanes" or "silicone". In addition the websites of regulating bodies in Europe, USA, and Australia were searched for reports related to PIP mammary implants. PIP mammary implants are more likely to rupture than other implants and can cause adverse effects in the short to the medium term related to the symptoms of rupture such as pain, lumps in the breast and axilla and anxiety. Based on peer-reviewed published studies we have calculated an overall rupture rate of 14.5% (383/2,635) for PIP implants. However, there is no evidence that PIP implant rupture causes long-term adverse health effects in humans so far. Silicone lymphadenopathy represents a foreign body reaction and should be treated conservatively. The long-term adverse effects usually arise from inappropriate extensive surgery, such as axillary lymph node dissection or extensive resection of breast tissue due to silicone leakage.

• 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.

• The Journal of the Korean dental association
• /
• v.53 no.2
• /
• pp.82-95
• /
• 2015

Background: Interest in medical malpractice claims and accidents is a day-to-day social issue to general public as well as medical personnel. Related laws and regulations already have been established, and institutions based on the laws and regulations also have been founded. However, in our dental community, interest and response to the issue seem insufficient. Methods: We searched four medical literature databases that are mainly cited in the medical community. Keywords including 'dental malpractice claims', 'patient safety' and 'medical accident' were used for the search. Among the selected literatures, we chose specific ones separately whose content is authentic and easily approachable. Results: Medical malpractice claims and accidents tend to increase around the world. As the cost or the difficulty level of surgery increases, the dispute rate also increases, which appears even more apparent in developed countries. Preventive measures to prevent the disputes and accidents are not significantly different. Three critical of them include relationship of doctor with patient, the informed consent and medical record. Conclusion: Tools for accident occurrence or communication improvement have been introduced. All of those cost time and money. However, education or professional request of liability insurance companies, self-education and provision of guidelines can be immediately implemented. To implement those, dentists' promotion at the regional or national level is imperative. rhBMP-2 is widely used at sinus augmentation, alveolar bone defect, and socket preservation.

• Journal of the Korean association of regional geographers
• /
• v.8 no.4
• /
• pp.415-429
• /
• 2002

The end of cold war and the trend of the detente brought about many changes in the military security policy and in the munitions industry of the West-Europe : reduction of the defence budget, diminution of the military power force and armaments, decrease of the exportation and the domestic demand of ordnance, and augmentation of discharged laborers. This mutation caused the munitions industry and the war industrials regions to a crisis. The munition industrials enterprises which was confronted a crisis searched for diverse strategies like an reconversion to the civil products, an internal diversification and an external growth. And the European Union, the national government, and local or regional self-government implemented policies for diversification of regional economic with an reconversion in order to minimize the effect of the regional economic and society from the crisis of the munition industry.