• Korean Journal of Culture and Social Issue
• /
• v.21 no.1
• /
• pp.21-43
• /
• 2015

The purpose of this study was to examine cultural differences and related difficulties which KoreanChinese women in Korean society experience at work and the different styles of their cultural adaptation. For this, semi-structured interviews consisting of questions related to these issues were conducted on 17 Korean-Chinese women living in the metropolitan area (average age: 34, SD = 9.25, average stay in Korea = 4 years, SD = 2.24). After analysis of qualitative data based on Giorgi's(1985) method of phenomenological analysis, a total of 225 significant statements were found and those were grouped into 23 subcategories, which were then grouped again into 9 categories. Cultural differences and related difficulties appeared in identity, verbal communication, political and economic aspects, and relational and sexual affairs. It was also revealed that Korean-Chinese women adapted in the three ways of Active, Passive, and Avoident. Lastly, various social actions that can aid the adaptation of Korean-Chinese women to Korea based on these results were discussed.

• Mood & Emotion
• /
• v.9 no.2
• /
• pp.78-86
• /
• 2011

Objectives : The aim of this study was to investigate relationship between perceived stress level, cognitive emotion regulation (CER) strategy and hope in group with high depressive symptom and higher-level perceived stress (H-H) and group with high depressive symptom and lower-level perceived stress (H-L) in patients with depression. Method : A total of 85 patients (over score of 16 by Beck depression Inventory; BDI) were surveyed with Cognitive emotion regulation questionnaire (CERQ), Perceived stress scale (PSS), and The state hope scale (SHS). Mean scores of CERQ and SHS were compared between relatively higher perceived stress and lower perceived stress groups. Correlation analysis and multiple linear regression analyses were performed to identify the effect of BDI, PSS and CER strategy on SHS in two groups. Results : In ANOVA, the level of hope and maladaptive CERQ score proved to be significantly lower among the H-H group than among the H-L group, while adaptive CERQ scores were not. In Regression analysis, the effective CER strategy in SHS were 'Refocus on planning' in H-H group, while it was 'Acceptance' CER strategy in H-L group. The final regression model explained 36% of the variance of SHS in H-H group and explained 21% of SHS in H-L group. Conclusion : These findings suggest that 'Refocus on planning' and 'Acceptance' cognitive emotion strategy are helpful in promotion of state hope on depression. Especially, 'Refocus on planning' strategy is more effective in high depressive symptom and high-perceived stress level, while 'Acceptance' strategy help to promote hope in high depressive symptom and low-perceived stress level in patients with depression.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3383-3397
• /
• 2023

Scene graphs serve as semantic abstractions of images and play a crucial role in enhancing visual comprehension and reasoning. However, the performance of Scene Graph Generation is often compromised when working with biased data in real-world situations. While many existing systems focus on a single stage of learning for both feature extraction and classification, some employ Class-Balancing strategies, such as Re-weighting, Data Resampling, and Transfer Learning from head to tail. In this paper, we propose a novel approach that decouples the feature extraction and classification phases of the scene graph generation process. For feature extraction, we leverage a transformer-based architecture and design an adaptive calibration function specifically for predicate classification. This function enables us to dynamically adjust the classification scores for each predicate category. Additionally, we introduce a Distribution Alignment technique that effectively balances the class distribution after the feature extraction phase reaches a stable state, thereby facilitating the retraining of the classification head. Importantly, our Distribution Alignment strategy is model-independent and does not require additional supervision, making it applicable to a wide range of SGG models. Using the scene graph diagnostic toolkit on Visual Genome and several popular models, we achieved significant improvements over the previous state-of-the-art methods with our model. Compared to the TDE model, our model improved mR@100 by 70.5% for PredCls, by 84.0% for SGCls, and by 97.6% for SGDet tasks.

• Journal of Internet Computing and Services
• /
• v.25 no.2
• /
• pp.1-9
• /
• 2024

Although federated learning is designed to be safer than centralized methods in terms of security and privacy, it still has many vulnerabilities. An attacker performing an adversarial attack intentionally manipulates the deep learning model by injecting carefully crafted input data, that is, adversarial examples, into the client's training data to induce misclassification. A common defense strategy against this is so-called adversarial training, which involves preemptively learning the characteristics of adversarial examples into the model. Existing research assumes a scenario where all clients are under adversarial attack, but considering the number of clients in federated learning is very large, this is far from reality. In this paper, we experimentally examine aspects of adversarial training in a scenario where some of the clients are under attack. Through experiments, we found that there is a trade-off relationship in which the classification accuracy for normal samples decreases as the classification accuracy for adversarial examples increases. In order to effectively utilize this trade-off relationship, we present a method to perform adversarial training by adaptively selecting a loss function depending on whether the client is attacked.

• Journal of Industrial Convergence
• /
• v.22 no.2
• /
• pp.63-71
• /
• 2024

Conventional methods like PIN used in conventional smartphone form factor have not considered the variation in display structure or screen size. As a result, when applied to recent variable display-based smartphones, the secret information input unit may get reduced or enlarged, leading to vulnerabilities for social engineering attacks due to deformation of the display area. This study proposes a secure keypad that responds to changes in display size in rollable and bendable smart phones. Firstly, the security problems that may arise when applying classical authentication methods to new form factors were analyzed, and corresponding security requirements were derived. The proposed security keypad addresses the key input error problem that can occur when the screen size is small. The arrangement and size of keys can be deformed with the spacing suitable for input depending on the display size of rollable and bendable smartphones. The study also considered the problem of leaking input information for social engineering attacks by irregularly distributing key input coordinates. The proposed method provides better user experience and security than existing methods and can be used in smartphones of various sizes and shapes.

• Journal of the Korean Society of Radiology
• /
• v.6 no.5
• /
• pp.357-363
• /
• 2012

The purpose of this study on head computed tomography scan corporate reorganization adaptive iteration algorithm using the statistical noise, and quality assessment, reduction of dose was evaluated. Head CT examinations do not apply ASIR group [A group], ASIR 50 applies a group [B group] were divided into examinations. B group of each 46.9 %, 48.2 %, 43.2 %, and 47.9 % the measured in the phantom research result of measurement of CT noise average were reduced more than A group in the central part (A) and peripheral unit (B, C, D). CT number was measured with the quantitive analytical method in the display-image quality evaluation and about noise was analyze. There was A group and difference which the image noise notes statistically between B. And A group was high so that the image noise could note than B group (31.87 HUs, 31.78 HUs, 26.6 HUs, 30.42 HU P<0.05). The score of the observer 1 of A group evaluated 73.17 on 74.2 at the result 80 half tone dot of evaluating by the qualitative evaluation method of the image by the bean curd clinical image evaluation table. And the score of the observer 1 of B group evaluated 71.77 on 72.47. There was no difference (P>0.05) noted statistically. And the inappropriate image was shown to the diagnosis. As to the exposure dose, by examination by applying ASIR 50 % there was no decline in quality of the image, 47.6 % could reduce the radiation dose. In conclusion, if ASIR is applied to the clinical part, it is considered with the dose written much more that examination is possible. And when examination, it is considered that it becomes the positive factor when the examiner determines.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.4
• /
• pp.63-68
• /
• 2010

The Color Management System in recent printers adopts ICC profiles for both monitors and printers. However, the ICC profile doesn't contain the characteristics of reproduced color on each monitor, because the color on each monitor is changed by user adjustment such as color temperature, brightness, and contrast adjustment. It is also depended on the backlight type and lifetime. As a result, unwanted color is reproduced on the printed paper, not like that on the monitor. To overcome the color difference between monitors and printers, it is needed to control the information of ICC profile. That is, first, the ICC profile is generated by the measurement of monitors having user set, then, through the CMS, the color on monitors can be produced on printed paper. However, it is difficult to apply the above system for normal users due to absence of measuring equipment and time consuming process. Therefore, this paper proposes a novel color matching technique based on the estimation of condition for each monitor having user set. The estimation is performed by a simple comparison visual test using a test image on printed paper and monitor. Then, the condition of monitor is applied to the ICC profile. As a result, the new ICC profile contains the color difference between user monitor and printer. The experimental results show the printed images using our proposed method have almost similar color with those on monitors.

• Journal of IKEEE
• /
• v.13 no.4
• /
• pp.116-124
• /
• 2009

Recently the CPU performance of modem chipsets or multimedia processors of mobile phone is as high as notebook PC. That is why mobile phone has been emerged as a leading ICON on the convergence of consumer electronics. The various applications of mobile phone such as DMB, digital camera, video telephony and internet full browsing are servicing to consumers. To meet all the demands the image quality has been increasingly important. Mobile phone is a portable device which is widely using in both the indoor and outside environments, so it is needed to be overcome to deteriorate image quality depending on environmental light source. Furthermore touch window is popular on the mobile display panel and it makes contrast loss because of low transmittance of ITO film. This paper presents the image enhancement algorithm to be embedded on image enhancement SoC. In contrast enhancement, we propose Clipped histogram stretching method to make it adaptive with the input images, while S-shape curve and gain/offset method for the static application And CIELCh color space is used to sunlight readability enhancement by controlling the lightness and chroma components which is depended on the sensing value of light sensor. Finally the performance of proposed algorithm is evaluated by using histogram, RGB pixel distribution, entropy and dynamic range of resultant images. We expect that the proposed algorithm is suitable for image enhancement of embedded SoC system which is applicable for the small-sized mobile display.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.433-438
• /
• 1999

This paper describes Active Noise Cancellation/Control(ANC) method that removes the information of the unnecessary noise and doesn't remove the informations of the necessary noise(warning sound, operating sound etc.) for the induced noise of the mechanical system. In this paper, the noise source Is axial fan, and the Feedback Active Noise control method that can effectively control BPF generated from the axial fan is used, and the Filtered-X LMS algorithm for adaptive algorithms is used. The experiments are executed for two case(propagating noise in the duct, emission noise for exterior free field). The part to be removed is BPF noise, and the band-pass filter not to effect to the other frequencies is used. Also, to investigate the effect of the noise reduction for human, we are compared with the results that are controlled for using Loudness before and after. As a results, we are certified that the BPF is decreased only and frequencies outside of BPF are not affected, and we acquire the reduction effects of 6.7 dB Loudness Level, then the frequency to be removed is controlled. Therefore, we can be certified that sound pressure as well as loudness can be effectively decreased for human sound quality

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.23 no.7
• /
• pp.9-15
• /
• 2019

The purpose of present study is to assess the maximum width of the surface cracks using the histogram analysis of image processing techniques in concrete structures. For this purpose, the concrete crack image is acquired by the camera. The image is Grayscale coded and Binary coded. After Binary coded image is Dilate and Erode coded, the image is then recognized as separated objects by applying Labeling techniques. Over time, dust and stains may occur naturally on the surface of concrete. The crack image of concrete may include shadows and reflections by lighting depending on a surrounding conditions. In general, concrete cracks occur in a continuous pattern and noise of image appears in the form of shot noises. Bilateral Blurring and Adaptive Threshold apply to the Grayscale image to eliminate these effects. The remaining noises are removed by the object area ratio to the Labeled area. The maximum numbers of pixels and its positions in the crack objects without noises are calculated in x-direction and y-direction by Histogram analysis. The widths of the crack are estimated by trigonometric ratio at the positions of the pixels maximum numbers for the Labeled objects. Finally, the maximum crack width estimated by the proposed method is compared to the crack width measured with the crack gauge. The proposed method by the present study may increase the reliability for the estimation of maximum crack width using image processing techniques in concrete surface images.