• Proceedings of the Korean Environmental Sciences Society Conference
• /
• 한국환경과학회 2020년도 정기학술대회 발표논문집
• /
• pp.216-216
• /
• 2020

• Proceedings of the Korea Multimedia Society Conference
• /
• 한국멀티미디어학회 2002년도 춘계학술발표논문집(하)
• /
• pp.553-558
• /
• 2002

본 논문에서는 컬러 영상에서 피부색 분할과 소벨 마스크를 이용한 얼굴 영역 검출 알고리즘을 제안한다. 제안된 알고리즘은 YCbCr색공간에서 Cb와 Cr성분을 이용하여 피부색 분할을 한 후에 형태학적 필터링과 레이블링을 통해 얼굴 후보 영역을 분리한다. 분리된 각 후보 영역에 대해 휘도 성분 Y에서 소벨 마스크의 수직 연산자를 적용한 후에 수평 투영을 통해 나타난 최대값을 눈의 위치로 검출해낸다. 비슷하게 얼굴의 지형적인 특징과 소벨 마스크의 수평 연산자를 적용하여 계산된 수평 투영의 최대값에 따라 턱 부분을 검출한다. 컴퓨터 시뮬레이션 결과는 제안된 방법이 기존의 방법보다 얼굴 영역을 정확하게 분리할 수 있음을 보인다.

• Proceedings of the Korean Information Science Society Conference
• /
• 한국정보과학회 2012년도 한국컴퓨터종합학술대회논문집 Vol.39 No.1(B)
• /
• pp.363-365
• /
• 2012

데이터의 규모 및 활용도, 그리고 사용자 접근성 측면에서 실세계 데이터에서 가장 중요한 이슈가 되는 것은 비디오 데이터이다. 장르나 등장인물, 배경 등이 매우 상이한 대량의 비디오 데이터들이 등장하고 있기 때문에, 통일된 사전지식을 이용한 비디오 데이터 분석이 매우 비현실적이 되어가고 있으며 사전지식을 활용하지 않는 비디오 분석기법의 중요성이 커지고 있다. 본 논문에서는 진화 파티를 필터링과 우점 이미지를 이용하여 비디오 데이터를 분절(Segmentation)하는 기법을 소개한다. 이미지 분절화 과정에서 해결해야 할 난점은 시점 변화 및 움직임 등에 의해 발생하는 사소한 변화가 컴퓨터 관점에서는 무시하기 어려운 큰 변화로 해석될 수 있다는 점이다. 동일장면에서의 시점 변화와 같은 사소한 변화로 인하여 동일 세그먼트를 추정하지 못하는 어려움을 해결하기 위하여 우리는 이미지 일부를 표현하는 파티클의 개체군을 생성하여 협력적인 방식으로 개별 이미지 세그먼트를 표현하는 방법을 개발하였다. 또한 동일 인물의 움직임과 같은 변화에 대응할 수 있도록 진화 파티를 필터링 방법을 컬러 히스토그램 방법과 결합하여 추론 성능을 한층 개선하였다. 실제 TV 드라마에 대하여 수행된 인간 평가자의 분절 평가 결과와 비교하여 제안 방법의 성능을 확인하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 한국컴퓨터정보학회 2023년도 제68차 하계학술대회논문집 31권2호
• /
• pp.21-22
• /
• 2023

딥러닝 모델은 영상 처리와 불량 감지 등 다양한 분야에서 높은 성능을 발휘하여 산업적으로 매우 중요하고 필수적인 기술이 되었다. 특히, 불량 감지는 제조업 분야에서 제품 품질 향상과 생산성 향상에 핵심적인 역할을 하는 기술로써 큰 관심을 받고 있다. 본 논문에서는 재건축 방식의 대표적인 모델인 DRAEM에 대해 불량 감지 성능을 향상하는 방법을 제시합니다. 이를 위해 컬러 분포의 차이를 최소화하는 손실 함수와 마스킹에 고주파 필터링을 적용하여 노이즈를 제거하는 방법을 활용한다. 이러한 방법들을 통해 DRAEM 모델의 성능을 개선하고, 정확하고 효과적인 불량 감지를 실현할 수 있다.

• Journal of Broadcast Engineering
• /
• 제16권6호
• /
• pp.962-968
• /
• 2011

The conventional lossless compression of color images is to apply a compression method to each of color components separately, without considering the channel correlation. There had been several methods that consider the channel correlation, but they were confined to the compression of satellite or aerial images only, and the performance of these algorithms to general photos is not satisfactory. This paper proposes a new lossless color image compression method that exploits the correlation between the color components. Specifically, asymmetric sampling is applied to transform an image into mosaic image and the rest, which are compressed separately. By using the information from the compressed mosaic image, the rest images are predicted for further reducing the information to be compressed. Experimental results show that the proposed method improves the compression performance by 35% over the conventional separate compression methods and 10% over the existing methods that exploit the channel correlation.

• The Journal of the Korea institute of electronic communication sciences
• /
• 제13권3호
• /
• pp.525-532
• /
• 2018

Recently, due to the advancements of sensor network technologies and camera technologies, there are increasing needs to effectively monitor the environment in a region that is difficult to access by using the visual sensor network that combines these two technologies. Since the image captured by the visual sensor reflects the natural phenomenon as it is, the quality of the image may deteriorate depending on the weather or time. In this paper, we propose an algorithm to improve the contrast of images using the characteristics of images obtained from visual sensors. In the proposed method, we first set the region of interest and then analyzes the change of the color value of the region of interest according to the brightness value of the image. The contrast of an image is improved by using the high contrast image of the same object and the analysis information. It is shown by experimental results that the proposed method improves the contrast of an image by restoring the color components of the low contrast image simply and accurately.

• The Journal of the Korea Contents Association
• /
• 제11권12호
• /
• pp.511-521
• /
• 2011

As the use of image input and output devices increases, the importance of extracting text area in color images is also increasing. In this paper, in order to extract text area of the images efficiently, we present a text area extraction method for color images based on labeling and gradient difference method. The proposed method first eliminates non-text area using the processes of labeling and filtering. After generating the candidates of text area by using the property that is high gradient difference in text area, text area is extracted using the post-processing of noise removal and text area merging. The benefits of the proposed method are its simplicity and high accuracy that is better than the conventional methods. Experimental results show that precision, recall and inverse ratio of non-text extraction (IRNTE) of the proposed method are 99.59%, 98.65% and 82.30%, respectively.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제21권10호
• /
• pp.454-460
• /
• 2020

In recent years, the rapid increases in video distribution and viewing over the Internet have increased the risk of personal information exposure. In this paper, a method is proposed to robustly identify areas in images where a person's privacy is compromised and simultaneously blocking the object area by blurring it while rapidly tracking it using a prediction algorithm. With this method, the target object area is accurately identified using artificial neural network-based learning. The detected object area is then tracked using a location prediction algorithm and is continuously blocked by blurring it. Experimental results show that the proposed method effectively blocks private areas in images by blurring them, while at the same time tracking the target objects about 2.5% more accurately than another existing method. The proposed blocking method is expected to be useful in many applications, such as protection of personal information, video security, object tracking, etc.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• 제42권3호
• /
• pp.73-82
• /
• 2005

Recently, The wide range of the Internet applications and the related technology developments enabled the ease use of the digital multimedia contents (fixed images, movies, digital audios). However, due to the replay ability which the contents may be easily duplicated and not only the duplicates are capable of providing the same original quality. There are mainly the encipher techniques and the watermarking techniques which are studied and used as solutions for the above problem in order to protect the license holders' rights. To the protection of the IP(Intellectual Property) rights of the owner, digital watermarking is the technique that authenticates the legal copyrighter. This paper proposed the watermarking algorithms to watermark the 256 gray logo image and the color image by applying the wavelet transformation to the color stand-still images. The proposed algorithms conducted the watermark insertion at the LH frequency region among the wavelet transformation regions (LL, LH, HL, HH). The interleaving algorithms which applied in data communication was applied to the watermark. the amount of watermark increased which consequently caused the PSNR to decrease but this might provide the perseverance against the external attacks such as extraction, filtering, and crop.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• 제11권3호
• /
• pp.847-852
• /
• 2010

Text in natural images has a various and important feature of image. Therefore, to detect text and extraction of text, recognizing it is a studied as an important research area. Lately, many applications of various fields is being developed based on mobile phone camera technology. Detecting edge component form gray-scale image and detect an boundary of text regions by local standard deviation and get an connected components using Euclidean distance of RGB color space. Labeling the detected edges and connected component and get bounding boxes each regions. Candidate of text achieved with heuristic rule of text. Detected candidate text regions was merged for generation for one candidate text region, then text region detected with verifying candidate text region using ectilarity characterization of adjacency and ectilarity between candidate text regions. Experctental results, We improved text region detection rate using completentary of edge and color connected component.