본 논문은 널리 알려진 RGB 색상 기반의 웹캠을 사용한 손 영역을 효율적으로 분할하는 방법을 제안한다. 이 방법은 잡음을 제거하기 위하여 네 번의 경험적 전처리 방법을 수행한다. 먼저, 전체 영상 잡음을 제거하기 위하여 가우시안평활화를 수행한다. 다음으로, RGB 영상은 HSV와 YCbCr 색상 모델로 변환되어, 각 색상 모델에 대해 통계적인 값에 기반하여 전역 고정 이진화가 수행된 후, 잡음은 bitwise-OR 연산에 의해 제거된다. 다음으로, 윤곽 근사화와 내부 영역 구멍 연산을 위해 RDP와 flood fill 알고리즘이 사용된다. 끝으로, 모폴로지 연산을 통하여 잡음을 제거하고 영상의 크기에 비례한 임계값을 결정하여 손 영역이 결정된다. 본 연구는 잡음 제거에 초점을 맞추고 있고 손 동작 인식 응용 기술에 사용될 수 있다.
본 논문은 널리 알려진 RGB 색상 기반의 웹캠을 사용한 손 영역을 효율적으로 분할하는 방법을 제안한다. 이 방법은 잡음을 제거하기 위하여 네 번의 경험적 전처리 방법을 수행한다. 먼저, 전체 영상 잡음을 제거하기 위하여 가우시안 평활화를 수행한다. 다음으로, RGB 영상은 HSV와 YCbCr 색상 모델로 변환되어, 각 색상 모델에 대해 통계적인 값에 기반하여 전역 고정 이진화가 수행된 후, 잡음은 bitwise-OR 연산에 의해 제거된다. 다음으로, 윤곽 근사화와 내부 영역 구멍 연산을 위해 RDP와 flood fill 알고리즘이 사용된다. 끝으로, 모폴로지 연산을 통하여 잡음을 제거하고 영상의 크기에 비례한 임계값을 결정하여 손 영역이 결정된다. 본 연구는 잡음 제거에 초점을 맞추고 있고 손 동작 인식 응용 기술에 사용될 수 있다.
This paper proposes a method for effectively segmenting the hand region using a widely popular RGB color-based webcam. This performs the empirical preprocessing method four times to remove the noise. First, we use Gaussian smoothing to remove the overall image noise. Next, the RGB image is converted...
This paper proposes a method for effectively segmenting the hand region using a widely popular RGB color-based webcam. This performs the empirical preprocessing method four times to remove the noise. First, we use Gaussian smoothing to remove the overall image noise. Next, the RGB image is converted into the HSV and the YCbCr color model, and global fixed binarization is performed based on the statistical value for each color model, and the noise is removed by the bitwise-OR operation. Then, RDP and flood fill algorithms are used to perform contour approximation and inner area fill operations to remove noise. Finally, ROI (hand region) is selected by eliminating noise through morphological operation and determining a threshold value proportional to the image size. This study focuses on the noise reduction and can be used as a base technology of gesture recognition application.
This paper proposes a method for effectively segmenting the hand region using a widely popular RGB color-based webcam. This performs the empirical preprocessing method four times to remove the noise. First, we use Gaussian smoothing to remove the overall image noise. Next, the RGB image is converted into the HSV and the YCbCr color model, and global fixed binarization is performed based on the statistical value for each color model, and the noise is removed by the bitwise-OR operation. Then, RDP and flood fill algorithms are used to perform contour approximation and inner area fill operations to remove noise. Finally, ROI (hand region) is selected by eliminating noise through morphological operation and determining a threshold value proportional to the image size. This study focuses on the noise reduction and can be used as a base technology of gesture recognition application.
* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.
문제 정의
The statistical value referred to here is a statistical value for skin color based on east Asian people[12]. Therefore, the hand region segmentation method proposed in this paper is based on the skin color of east Asian people.
제안 방법
Next, global fixed binarization is performed using HSV and YCbCr color model, and bitwise-OR operation is performed to remove noise. Then, the contour approximation is performed using the RDP (Ramer Douglas Peuker) algorithm and the noise is removed by filling the approximated contour with the flood fill algorithm. Finally, the noise is removed by morphology operation, and the hand region is selected by determining the image size proportional threshold value.
Therefore, it is recommended not to use frame rate and exposure compensation function as much as possible to obtain stable image information. In addition, the method proposed in this paper includes a global fixed binary technique using statistical numerical values. The statistical value referred to here is a statistical value for skin color based on east Asian people[12].
이론/모형
First, the overall noise is removed for the real-time frame obtained from the webcam. The Gaussian smoothing technique is applied to this technique, and the noise is randomly distributed, and the Gaussian distribution is observed. That is, a convolution is performed on the input image using a 15 × 15 Gaussian kernel generated through a two-dimensional Gaussian function such as eq.
성능/효과
11, the results of experiment with 105 frame images in a relatively uniform illumination environment are shown in Table 2. The test was conducted in an environment with a small change in luminance and uniform illumination, showing an accuracy of 86.8517%. The accuracy of the test for most frame images was more than 98%, but the accuracy was reduced by about 70 ~ 80% due to the reflected light from the hand or the afterglow when the hand was moved quickly.
8517%. The accuracy of the test for most frame images was more than 98%, but the accuracy was reduced by about 70 ~ 80% due to the reflected light from the hand or the afterglow when the hand was moved quickly. Therefore, unlike the experimental data presented in Table 2, the actual arithmetic mean of 86.
Compared with the conventional single color model based hand area segmentation method, noise reduction and hand region segmentation accuracy are improved. Experimental results of 105 frame images extracted randomly from RGB color based real time images show 86.8517%. Therefore, it can be used as a base technology of a contactless gesture interface using a RGB color-based webcam in a general real-life environment.
참고문헌 (15)
D. Hong and W. Woo, "Recent Research Trend of Gesture-based User Interfaces", Telecommunications review, Vol. 18(3), pp. 403-413, 2008.
S. Park and E. Lee, "Hand Gesture Recognition Algorithm Robust to Comple Image", Journal of Korea Multimedia Society, Vol. 13, pp. 1000-1015, 2010.
V. Vezhnevets, V. Sazonov and A. Andreeva, "A Survey on Pixel-Based Skin Color Detection Techniques", Proc. Graphicon, Vol. 3, pp. 85-92, 2003.
M. Na, H. Kim and T. Kim, "An Illumination and Background-Robust Hand Image Segmentation Method Based on the Dynamic Threshold Values", Journal of Korea Multimedia Society, Vol. 14(5), pp. 607-613, 2011.
F. Gasparini and R. Schettini, "Skin Segmentation using multiple thresholding", Proceedings of SPIE, Vol. 6061, pp. 1-8, 2006.
M. Jones and J. M. Rehg, "Statistical Color Models with Application to Skin Detection", International Journal of Computer Vision, Vol. 46, pp. 81-96, 2002.
S. L. Phung, A. Bouzerdoum and D. Chai, "Skin Segmentation Using Color Pixel Classification:Analysis and Comparison", IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 27, pp. 148-154, 2005.
J. Han, G. M. Award, A. Sutherland and H. Wu, "Automatic Skin Segmentation for Gesture Recognition Combining Region and Support Vector Machine Active Learning", Proceedings of the 7th International Conference on Automatic Face and Gesture Recognition, pp. 57-64, 2006.
M. Storring, T. Kocka, H. J. Andersen and E. Granum, "Tracking Regions of Human Skin through Illumination Changes", Elsevier Pattern Recognition Letters, Vol. 24, pp. 1715-1723, 2003.
P. Kakumanu, S. Makrogiannis and N. Bourbakis, "A Survey of Skin-Color Modeling and Detection Methods", Elsevier Pattern Recognition, Vol. 40, pp. 1106-1122, 2007.
G. D. Finlayson, B. Schiele, and J. L. Crowley, "Comprehensive Colour Image Normalization", Proc. European Conference on Computer Vision, Vol. 1, pp. 475-490, 1998.
S. Cho, J. Bae and S. Lee, "Adaptive Skin Segmentation based on Region Histogram of Color Quantization Map", Journal of KIISE: Software and Applications, Vol. 36(1), pp. 54-61, 2009.
J. Kim and M. Kang, "A Study on Hand Shape Recognition using Edge Orientation Histogram and PCA", Journal of Digital Contents Society, Vol. 10, pp. 319-326, 2009.
S. Kang, M. Nam and P. Rhee, "Color Based Hand and Finger Detection Technology for User Interaction", International Conference on Convergence and Hybrid Information Technology, pp. 229-236, 2008.
J. G. Back and etc, "Extracting the Slope and Compensating the Image Using Edges and Image Segmentation in Real World Image", Journal of DCS, Vol. 17, No. 5, pp. 441-448, Oct. 2016
이 논문을 인용한 문헌
저자의 다른 논문 :
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
활용도 분석정보
상세보기
다운로드
내보내기
활용도 Top5 논문
해당 논문의 주제분야에서 활용도가 높은 상위 5개 콘텐츠를 보여줍니다. 더보기 버튼을 클릭하시면 더 많은 관련자료를 살펴볼 수 있습니다.
AI-Helper
※ AI-Helper는 부적절한 답변을 할 수 있습니다.