• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.18 no.8
• /
• pp.2333-2345
• /
• 2024

Facial wrinkles are widely used to evaluate skin condition or aging for various fields such as skin diagnosis, plastic surgery consultations, and cosmetic recommendations. In order to effectively process facial wrinkles in facial image analysis, accurate wrinkle segmentation is required to identify wrinkled regions. Existing deep learning-based methods have difficulty segmenting fine wrinkles due to insufficient wrinkle data and the imbalance between wrinkle and non-wrinkle data. Therefore, in this paper, we propose a new facial wrinkle segmentation method based on a UNet++ model. Specifically, we construct a new facial wrinkle dataset by manually annotating fine wrinkles across the entire face. We then extract only the skin region from the facial image using a facial landmark point extractor. Lastly, we train the UNet++ model using both dice loss and focal loss to alleviate the class imbalance problem. To validate the effectiveness of the proposed method, we conduct comprehensive experiments using our facial wrinkle dataset. The experimental results showed that the proposed method was superior to the latest wrinkle segmentation method by 9.77%p and 10.04%p in IoU and F1 score, respectively.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.1
• /
• pp.436-442
• /
• 2017

A novel method of Skin lesion segmentation based on the combination of Texture and Neural Network is proposed in this paper. This paper combines the textures of different pixels in the skin images in order to increase the performance of lesion segmentation. For segmenting skin lesions, a two-step process is done. First, automatic border detection is performed to separate the lesion from the background skin. This begins by identifying the features that represent the lesion border clearly by the process of Texture analysis. In the second step, the obtained features are given as input towards the Recurrent Echo state neural networks in order to obtain the segmented skin lesion region. The proposed algorithm is trained and tested for 862 skin lesion images in order to evaluate the accuracy of segmentation. Overall accuracy of the proposed method is compared with existing algorithms. An average accuracy of 98.8% for segmenting skin lesion images has been obtained.

• Proceedings of the Korean Institute of Interior Design Conference
• /
• 2005.10a
• /
• pp.165-170
• /
• 2005

It is important factor; understand definition and concept, grasp application method and property about transparency for expression of skin in design of retail space. This research do target; clarify the feature of transparency for expression of skin in modern retail space, and it is based in these viewpoint that analyze the feature through an experiment of image estimation(SD method) into object to modern retail space that express transparency of skin.

• Proceedings of the KIEE Conference
• /
• 2006.04a
• /
• pp.39-41
• /
• 2006

It is important to estimate the hand skin temperature because it reveals not only physiological properties of a certain diseases but also it can estimate even human mental-stress conditions. In this study, we try to estimate the temporal skin temperature distribution of human hand by applying stress-cold test to possibly apply to estimate a subject's blood circulation condition in his or her hand in terms of normal or abnormal state.

• Korean Institute of Interior Design Journal
• /
• no.36
• /
• pp.14-21
• /
• 2003

The Propose of this research was to consider how the New Media Architecture was applied to contemporary architecture according to the analysis of the design method process of Toyo Ito. Sendai Mediatheque by Toyo Ito stands as one of the most symbolic statement In New Media architecture. The four principal architectural elements of the Mediatheque are the digital image, the continuous space, the tube, and the skin facade. The digital image express forms of communication, person-to-person and person-to-thing, and they vary according to the media utilized on each level. The three skin elements of the Mediatheque are a double skin of MPG, skin of louvers, skin of fine-floor decking. The tubes act as columns while enveloping light, air, water, electricity the passage of people, as well as the means of transferring material. The thirteen tubes of different sizes prevent the erection of wall and suggest places instead of rooms. Instead of being limited to certain specified actions in clearly defined rooms, people are free to choose places for their actions in the continuous space.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.14 no.4
• /
• pp.69-77
• /
• 2018

The malignant melanoma accounts for about 1 to 3% of the total malignant tumor in the West, especially in the US, it is a disease that causes more than 9,000 deaths each year. Generally, skin lesions are difficult to detect the features through photography. In this paper, we propose a computer-aided diagnosis algorithm based on deep learning for classification of malignant melanoma and benign skin tumor in RGB channel skin images. The proposed deep learning model configures the tumor lesion segmentation model and a classification model of malignant melanoma. First, U-Net was used to segment a skin lesion area in the dermoscopic image. We could implement algorithms to classify malignant melanoma and benign tumor using skin lesion image and results of expert's labeling in ResNet. The U-Net model obtained a dice similarity coefficient of 83.45% compared with results of expert's labeling. The classification accuracy of malignant melanoma obtained the 83.06%. As the result, it is expected that the proposed artificial intelligence algorithm will utilize as a computer-aided diagnosis algorithm and help to detect malignant melanoma at an early stage.

• The Journal of Korean Society for Radiation Therapy
• /
• v.20 no.1
• /
• pp.17-23
• /
• 2008

Purpose: Cone-beam CT using linear accelerator attached to on-board imager is a image guided therapy equipment. Because it is to check the patient's set-up error, correction, organ and target movement. but imaging dose should be cause of the secondary cancer when taking a image. The aim of this study is investigation of appropriate cone beam CT scan mode to compare and estimate the image quality and skin dose. Materials and Methods: Measurement by Thermoluminescence dosimeter (TLD-100, Harshaw) with using the Rando phantom are placed on each eight sites in seperately H&N, thoracic, abdominal section. each 4 methods of scan modes of are measured the for skin dose in three time. Subsequently, obtained average value. Following image quality QA protocol of equipment manufacturers using the catphan 504 phantom, image quality of each scan mode is compared and analyzed. Results: The results of the measured skin dose are described in here. The skin dose of Head & Neck are measured mode A: 8.96 cGy, mode B: 4.59 cGy, mode C: 3.46 cGy mode D: 1.76 cGy and thoracic mode A: 9.42 cGy, mode B: 4.58 cGy, mode C: 3.65 cGy, mode D: 1.85 cGy, and abdominal mode A: 9.97 cGy, mode B: 5.12 cGy, mode C: 4.03 cGy, mode D: 2.21 cGy. Approximately, dose of mode B are reduced 50%, mode C are reduced 60%, mode D are reduced 80% a point of reference dose of mode A. the results of analyzed HU reproducibility, low contrast resolution, spatial resolution (high contrast resolution), HU uniformity in evaluation item of image quality are within the tolerance value by recommended equipment manufacturer in all scan mode. Conclusion: Maintaining the image quality as well as reducing the image dose are very important in cone beam CT. In the result of this study, we are considered when to take mode A when interested in soft tissue. And we are considered to take mode D when interested in bone scan and we are considered to take mode B, C when standard scan. Increasing secondary cancer risk due to cone beam CT scan should be reduced by low mAs technique.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.49 no.4
• /
• pp.10-15
• /
• 2012

In a color image, people and especially facial patterns are important and interesting visual objects. Thus, effective skin color reproduction is essential, as skin color is a key memory color in color application systems. Previous studies suggested skin color reproduction by mapping only to the center value of preferred skin region. However, it is not suitable to determine one preference color because preference color from the observer's preference test is not dominant. In this paper, skin color reproduction using multiple preferred skin colors for each race is proposed. The proposed method first defines multiple preferred skin colors for each race according to their luminance level. After that, skin region is detected in an image. The race is then selected by calculating distance between average chromaticity of detected region and that of each racial skin from a database to assign preferred skin color for each race. Next, each corresponding preferred skin color is determined for each selected race. Finally, input skin color is proportionally mapped toward preferred skin color according to the difference between the input skin color and the preferred skin color for a smoothly reproduced skin color. In the experimental results, the proposed method gives better color correction on the objective and subjective evaluation than the previous methods.

• Proceedings of the IEEK Conference
• /
• 2006.06a
• /
• pp.795-796
• /
• 2006

Recently, blocking harmful contents such as adult images has been widely researched. However, most of adult image blocking methods use flush color model without shape information. In this paper, we present a new adult image blocking methods based on shape and color information. In the first step, the shape is considered. In the second step, adult images are detected using skin color model. Considering both shape and color, the detection rate can be increased. We evaluate adult image detection performance using sample images.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.2
• /
• pp.1414-1419
• /
• 2015

A method for detecting a skin region on the image is generally used to detect the color information. However, If saturation lowered, skin detection is difficult because hue information of the pixels is lost. So in this paper, we propose a method of correcting color of lower saturation of skin region images by the lighting. Color correction process of this method is saturation image acquisition and low-saturation region classification, segmentation, and the saturation of the split in the low saturation region extraction and color values, the color correction sequence. This method extracts the low saturation regions in the image and extract the color and saturation in the region and the surrounding region to produce a color similar to the original color. Therefore, the method of extracting the low saturation region should be correctly preceding. Because more accurate segmentation in the process of obtaining a low saturation regions, we use a multi-threshold method proposed Otsu in Hue values of the HSV color space, and create a binary image. Our experimental results for 170 portrait images show a possibility that the proposed method could be used efficiently preprocessing of skin color detection method, because the detection result of proposed method is 5.8% higher than not used it.