• Smart Media Journal
• /
• v.13 no.3
• /
• pp.53-61
• /
• 2024

This paper presents a novel M-shaped encoder-decoder architecture for skin lesion segmentation, achieving better performance than existing approaches. The proposed architecture utilizes the left and right legs to enable multi-scale feature extraction and is further enhanced by integrating an attention module within the skip connection. The image is partitioned into four distinct patches, facilitating enhanced processing within the encoder-decoder framework. A pivotal aspect of the proposed method is to focus more on critical image features through an attention mechanism, leading to refined segmentation. Experimental results highlight the effectiveness of the proposed approach, demonstrating superior accuracy, precision, and Jaccard Index compared to existing methods

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.4
• /
• pp.19-30
• /
• 2009

In this paper, We propose a method with which we can detect facial components and face in input image. We use eye map and mouth map to detect facial components using eyes and mouth. First, We find out eye zone, and second, We find out color value distribution of skin region using the color around the eye zone. Skin region have characteristic distribution in YCbCr color space. By using it, we separate the skin region and background area. We find out the color value distribution of the extracted skin region and extract around the region. Then, detect mouth using mouthmap from extracted skin region. Proposed method is better than traditional method the reason for it comes good result with accurate mouth region.

• Journal of Digital Contents Society
• /
• v.19 no.7
• /
• pp.1397-1403
• /
• 2018

Today, modern people perceive skin care as part of their physical health care, and acne is a common skin disease problem that is found on the face. In this paper, an effective acne detection algorithm using CIE $L^*a^*b^*$ color space has been proposed. It is red when the pixel value of the component image $a^*$ is a positive number, so it is suitable for detecting acne in skin image. First, the skin image based on the RGB color space is subjected to light compensation through color balancing, and converted into a CIE $L^*a^*b^*$ color space. The extracted component image $a^*$ was normalized, and then the skin and acne area were estimated with the threshold values. Experimental results show that the proposed method detects acne more effectively than the conventional method based on brightness information, and the proposed method is robust against the reflected light source.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2021.05a
• /
• pp.551-552
• /
• 2021

본 논문에서는 촬영한 피부 영상(얼굴, 손 등)을 이용한 생체신호(맥박, 호흡, 혈압, 체온 등) 측정 S/W 기술을 제안한다. 기존의 생체신호 측정 기술은 다양한 센서(PPG, 압력 센서, 혈압계, 체온계 등)가 탑재된 측정 장치를 이용하여 상태를 측정하고 이를 진단하는 연구들이 진행되어 왔다. 각 각의 생체신호를 측정하기 위해서는 별도로 구비된 측정 장치들을 이용하여 개별적으로 생체신호를 측정하고 확인하여야 한다. 제안된 기술은 스마트 디바이스에 생체신호 측정 S/W의 설치만으로 카메라로 촬영한 피부 영상의 피부 관심 영역에서 계산된 색상 데이터를 이용하여 다양한 생체신호를 언제 어디서나 실시간으로 측정할 수 있으며, 생체신호 측정 성능 평가 결과 맥박수 2.63%, 호흡수 5.98%, 이완기 혈압 2.48%, 수축기 혈압 5.23% 및 체온 0.25%의 오차율이 계산되었다.

• Proceedings of the Optical Society of Korea Conference
• /
• 2009.02a
• /
• pp.241-242
• /
• 2009

도플러 효과를 이용해 피부의 혈류 속도 분포를 측정하였다. 고속 CMOS 이미지 센서를 사용하여 $256{\times}256$ 픽셀의 크기를 갖는 혈류 속도 영상을 측정하였다. 많은 의학 치료 분야에서 혈류 흐름에 대한 영상, 특히 모세혈관에서의 혈류 흐름에 대한 영상을 필요로 한다. 레이저 도플러 기술은 물체의 속도를 물체에 접촉하지 않고 측정할 수 있는 대표적인 기술로, 레이저 광의 코헤런트한 특성과 비접촉성 특성은 인체 피부의 혈류 속도를 측정하는데 좋은 특성을 제공한다.

• Proceedings of the KIEE Conference
• /
• 2011.07a
• /
• pp.2037-2038
• /
• 2011

피부에 나타나는 난치성 질환인 백반증은 심리적인 위축감을 주어 정상적인 생활에 지장을 줄 수 있는 질병이다. 이에 따라서 본 연구에서는 피부에 나타나는 백반증의 진행 상태를 판단하기 위하여 L*a*b* 컬러 공간으로 변환된 피부 영상에 Otsu 임계값 설정 기법을 적용하여 백반증의 발병 영역을 자동으로 판별하는 알고리즘을 제안하였다.

• Journal of the Korean Society of Radiology
• /
• v.81 no.3
• /
• pp.714-718
• /
• 2020

Herpetic mastitis is extremely rare, and its imaging findings remain unclear. We report a case of herpes simplex mastitis in an adolescent woman and describe the clinical and ultrasound features. The patient showed unilateral nipple and areolar skin thickening and axillary lymphadenopathy on B-mode ultrasonography. Doppler ultrasonography revealed multiple linear and branching blood flows in the areolar area. The lesion was verified as herpes simplex mastitis via a skin biopsy. This report shows that the radiologic features of herpes simplex mastitis may be similar to those of Paget's disease because of localized nipple and areolar skin thickening and increased vascularity.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.146-149
• /
• 2009

Face detection and recognition in real-time video constitutes one of the recent topics in the field of computer vision. In this paper, we propose face detection and tracking algorithm using the skin color and haar-like feature in real-time video sequence. The proposed algorithm further includes color space to enhance the result using haar-like feature and skin color. Experiment results reveal the real-time video processing speed and improvement in the rate of tracking.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.1
• /
• pp.35-44
• /
• 2010

In this paper, we propose a method about producing skin color model adaptively in input image and face detection. The principle process which we proposed is finding eyes candidates by applying the eye features to neural network, and then using the around color to find the distribution of color value. There will be a verification process that producing face region by using color value distribution which is detected as skin region and find mouth candidate in corresponding face region; if eye candidate and mouth candidate's connection structure is similar with face structure, then it can be judged as a face. Because this method can detect skin region adaptively by finding eyes, we solve the rate of false positive about the distorted skin color which is used by existing face detection methods. The experiment was performed about detecting the eye, the skin, the mouth and the face individually. The results revealed that the proposed technique is better than the traditional techniques.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.6
• /
• pp.29-36
• /
• 2012

In this paper, we suggest a new method of detecting human skin-color regions from three-dimensional static or dynamic stereoscopic images by effectively integrating depth and color features. The suggested method first extracts depth information that represents the distance between a camera and an object from input left and right stereoscopic images through a stereo matching technique. It then performs labeling for pixels with similar depth features and determines the labeled regions having human skin color as actual skin color regions. Our experimental results show that the suggested skin region extraction method outperforms existing skin detection methods in terms of skin-color region extraction accuracy.