• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.6
• /
• pp.1087-1098
• /
• 2023

The rapid advancement of artificial intelligence (AI) technology has led to its proactive utilization across various fields. However, this widespread adoption of AI-based systems has raised concerns about the increasing threat of attacks on these systems. In particular, deep neural networks, commonly used in deep learning, have been found vulnerable to adversarial attacks that intentionally manipulate input data to induce model errors. In this study, we propose a method to protect image classification models from visually imperceptible One-Pixel attacks, where only a single pixel is altered in an image. The proposed defense technique utilizes an autoencoder model to remove potential threat elements from input images before forwarding them to the classification model. Experimental results, using the CIFAR-10 dataset, demonstrate that the autoencoder-based defense approach significantly improves the robustness of pretrained image classification models against One-Pixel attacks, with an average defense rate enhancement of 81.2%, all without the need for modifications to the existing models.

• KIPS Transactions on Computer and Communication Systems
• /
• v.11 no.6
• /
• pp.185-192
• /
• 2022

Multimodal medical image fusion (MMIF) fuses two images containing different structural details generated in two different modes into a comprehensive image with saturated information, which can help doctors improve the accuracy of observation and treatment of patients' diseases. Therefore, a method based on double-layer decomposer and fine structure preservation model is proposed. Firstly, a double-layer decomposer is applied to decompose the source images into the energy layers and structure layers, which can preserve details well. Secondly, The structure layer is processed by combining the structure tensor operator (STO) and max-abs. As for the energy layers, a fine structure preservation model is proposed to guide the fusion, further improving the image quality. Finally, the fused image can be achieved by performing an addition operation between the two sub-fused images formed through the fusion rules. Experiments manifest that our method has excellent performance compared with several typical fusion methods.

• Culinary science and hospitality research
• /
• v.9 no.4
• /
• pp.163-178
• /
• 2003

The purpose of this study is to find the effect of restaurant brand image upon service quality, price acceptability, and revisit intention, and to propose the importance of brand image to operators and managers who manage restaurants. To accomplish the purpose of this study, sampling was taken among customers who visit six deluxe hotels and six family restaurants in Seoul. Six hundreds questionnaires were distributed to each hotel and restaurant and 487 valid samples were selected for statistical analysis. The questionnaire consists of 77 items about demographical characteristics, brand image, service quality, revisit intention, price acceptability, and spending patterns. SPSS WIN 10.0 was used for statistical analysis. A research model was built up and three null hypotheses were established. Based on theses research model and three null hypotheses, the test was conducted, and the results are as follows. Brand image has an effect upon service quality, and furthermore this can be preceding variable of service quality. Also Service quality has an effect upon price acceptability and revisit intention.

• Korean Journal of Construction Engineering and Management
• /
• v.9 no.5
• /
• pp.75-85
• /
• 2008

The construction industry has been leading the growth of the nation's economy not only by providing with various infrastructure projects but also by positively impaction related industries such as crating numerous job opportunities. Relevant statistics show the production amount of construction taking about 17.5% of the GDP (Gross Domestic Product). In spite of its positive impacts on the economy, image of Korea construction industry is generally reflected as negative mainly because of its environmental disruption, low payment, bribe, fraudulent work and inefficiency. It brings students to be reluctant choosing the construction industry as their carrier path and governmental and principal research status. Therefore it has been difficult to recruit highly qualitied human resources to the industry while the morale of the whole industry has gradually become demoralized. To improve this stand, many domestic researchers carried out research projects for improving the image of Korea Construction Industry. This study also sympathizes with necessity of improving the negative image of construction industry to remain as one of the leading industry in the 21st century. Especially, this study focused on finding important factors which have significant influences on the image of the industry. Through out the research, image influence factors was identified by rigorous literature review and interviews as industrial and academic experts. Factors, then, categorized and used as the main framework for the survey which designed to fine the degree of impacts on the image of the construction industry. In analyzing the survey results, various statistical techniques was employed including factor analysis, Chi-Square-Test, Correlation Analysis and Multiple Linear Regression. Identified as the most influent factors to the image of the construction industry include morale of construction employee, and prospects the industry which of the judgement by payment, impacts on nation's economy, future of the industry, etc.

• Journal of Korean Society of Forest Science
• /
• v.90 no.6
• /
• pp.747-755
• /
• 2001

This study was analyzed the destination image and tests of image value of forest region festival by segmenting the market. A field survey was conducted during the Acasia-Honey Festival days in Chilgok, the famous acasia-honey products place, from May 12-15, 2001. In this study, 20 image items on five-point Likert scale were examined through factor analysis using a sample of 394 visitors. Four factors were identified as follows : environment, behavior-facilities, festival programs, and products. The author observed if there were statistically significant differences of destination images between the first-time and the repeat visitors, between residents and non-residents, and between purchasers and non-purchasers of acasia-honey products. The results indicated that repeat visitors and residents group had a more positive image of behavior factor such as hospitality and information for festival. And, the author found that purchasers of acasia-honey products had more positive image of product as well as environment factor such as special atmosphere, nostalgia, natural attractions, local culture, and novelty. For these results, it is possible to conclude that local government must effectively provide the festival visitors with well-defined information such as festival contents and schedule, also, develop the programs involved unique attributes of forest region festival to improve festival image.

• Journal of Broadcast Engineering
• /
• v.27 no.1
• /
• pp.69-79
• /
• 2022

In this study, a method of generating and utilizing a normal map image used to represent the characteristics of the surface of an image material to improve the classification accuracy of the original material image is proposed. First of all, (1) to generate a normal map that reflects the surface properties of a material in an image, a U-Net with attention-R2 gate as a generator was used, and a Pix2Pix-based method using the generated normal map and the similarity with the original normal map as a reconstruction loss was used. Next, (2) we propose a network that can improve the accuracy of classification of the original material image by applying the previously created normal map image to the attention gate of the classification network. For normal maps generated using Pixar Dataset, the similarity between normal maps corresponding to ground truth is evaluated. In this case, the results of reconstruction loss function applied differently according to the similarity metrics are compared. In addition, for evaluation of material image classification, it was confirmed that the proposed method based on MINC-2500 and FMD datasets and comparative experiments in previous studies could be more accurately distinguished. The method proposed in this paper is expected to be the basis for various image processing and network construction that can identify substances within an image.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2010.05a
• /
• pp.371-372
• /
• 2010

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2009.06a
• /
• pp.333-334
• /
• 2009

• Proceedings of the Korean Institute of Landscape Architecture Conference
• /
• 2005.10a
• /
• pp.48-51
• /
• 2005

• Proceedings of the Speleological Society Conference
• /
• 2001.11a
• /
• pp.9-20
• /
• 2001

국민소득의 향상과 여가생활의 증대로 관광산업은 환경산업과 더불어 미래의 주요산업으로 인식되고 있다. 자연동굴은 신비스러운 내부 경관과 자연생태의 보고라는 교육적 관점에서 매우 훌륭한 관광자원으로 그 가치를 인정받고 있으며 최근에는 자연동굴을 보유하고 있는 지방자치단체의 경우 지역개발 및 관광수입원으로서 자연동굴의 개발에 많은 관심을 보이고 있다. (중략)