• Journal of the Korea institute for structural maintenance and inspection
• /
• v.26 no.2
• /
• pp.28-36
• /
• 2022

As port structures are exposed to various extreme external loads such as wind (typhoons), sea waves, or collision with ships; it is important to evaluate the structural safety periodically. To monitor the port structure, especially the rubber fender, a fender segmentation system using a vision sensor and deep learning method has been proposed in this study. For fender segmentation, a new deep learning network that improves the encoder-decoder framework with the receptive field block convolution module inspired by the eccentric function of the human visual system into the DenseNet format has been proposed. In order to train the network, various fender images such as BP, V, cell, cylindrical, and tire-types have been collected, and the images are augmented by applying four augmentation methods such as elastic distortion, horizontal flip, color jitter, and affine transforms. The proposed algorithm has been trained and verified with the collected various types of fender images, and the performance results showed that the system precisely segmented in real time with high IoU rate (84%) and F1 score (90%) in comparison with the conventional segmentation model, VGG16 with U-net. The trained network has been applied to the real images taken at one port in Republic of Korea, and found that the fenders are segmented with high accuracy even with a small dataset.

• Smart Media Journal
• /
• v.11 no.11
• /
• pp.17-24
• /
• 2022

With the spread of smart speakers based on voice recognition technology and deep learning technology, not only non-disabled people, but also the blind or physically handicapped can easily control home appliances such as lights and TVs through voice by linking home network services. This has greatly improved the quality of life. However, in the case of speech-impaired people, it is impossible to use the useful services of the smart speaker because they have inaccurate pronunciation due to articulation or speech disorders. In this paper, we propose a personalized voice classification technique for the speech-impaired to use for some of the functions provided by the smart speaker. The goal of this paper is to increase the recognition rate and accuracy of sentences spoken by speech-impaired people even with a small amount of data and a short learning time so that the service provided by the smart speaker can be actually used. In this paper, data augmentation and one cycle learning rate optimization technique were applied while fine-tuning ResNet18 model. Through an experiment, after recording 10 times for each 30 smart speaker commands, and learning within 3 minutes, the speech classification recognition rate was about 95.2%.

• Korean Security Journal
• /
• no.31
• /
• pp.157-185
• /
• 2012

Today, the emergence of cyberspace and advancement of globalization caused not only the transformation of our productive and conventional life but also the revolutionary transition of use of destructive violence such as crime and warfare. This transition of environmental condition connects various security threats which separatedly existed in individual, local, national, and global levels in the past, and transformed the mechanical sum of all levels of security threats into the organic sum of multi-dimensional security threats. This article proposes that the sum of multi-dimensional security threats is caused by the interconnectivity of various different levels of security threats and the integrated interdisciplinary perspective is essential to properly understand the fundamental existence of today's security problem and the reality of fear that we face today. The holistic security, the concept proposed here, is to suggest the mode of networked response to multi-dimensional security threats. The holistic security is suggested to overcome the conventional divisional approach based on the principle of "division of labor" and bureaucratic principles, which means more concretely that national security and criminal justice are divided and intelligence, military, police, prosecution, fire-fighting, private security, and etc. are strictly separated into its own expertise and turf. Also, this article introduces integrated security approaches tried by international organization and major countries overseas with the respect of the holistic security. The author have spent some substantial experience of participant observation, meetings, seminar, conference, and expert interviews regarding the issues discussed in the article in various countries including the United States, Russia, Austria, Germany, Canada, Mexico, Israel, and Uzbekistan for the last ten years. Intelligence and information on various levels of security threats and security approaches introduced in this paper is obtained from such opportunities.

• The Journal of the Korean bone and joint tumor society
• /
• v.15 no.1
• /
• pp.34-43
• /
• 2009

Purpose: Glomus tumors are rare benign vascular tumors, usually located in the skin or soft tissue of extremities. Approximately 30-50% of glomus tumors occur in subungal area, but glomus tumors have been described in every location even where glomus bodies are not or rarely present. The purpose of this study was to identify clinical, histologic and MRI characteristics of soft tissue glomus tumor. Materials and Methods: Between 1993 and 2008, eight patients underwent surgery of soft tissue Glomus tumor at our institution. Exclusion criteria were patients with Glomus tumors in digits, stomach, trachea and glomus tympanicum. We analyzed medical records, interviews, physical examinations, MR findings and histolocial types retrospectively. Results: There were four men and four women. The mean age was fourty-seven years. The mean prevalence time was eight-point-nine years. In the classic triad of symptoms, all eight patients had pain and tenderness. Two patients complained of cold sensitivity. Two showed skin color changes. After surgery, two showed symptom improvement (VAS $9^{\circ}{\rightarrow}8$, $8^{\circ}{\rightarrow}5$) and?six showed complete disappearance of symptoms. Slightly symptom improvemented (VAS $9^{\circ}{\rightarrow}8$) one had additional surgery two times after first surgery due to relapse/remaining Glomus tumor. The mean size was 13.9 mm. In histology, six were 'solid glomus tumor', one was a mixture of 'solid glomus tumor' and 'lomangioma' and one was 'malignant glomus tumor'. MR findings showed isointense signal on T1 image, high signal on T2 image and strong enhancement on the Gadolinium enhanced image. Conclusion: Glomus tumor has low recurrence rate and malignant change, rapid diagnosis and surgical excision is critical in treatment to prevent unnecessary pain of patient.