Blucher Engineering Proceedings

Novembro 2019, vol.6, num.3 - V Simpósio Internacional de Inovação e Tecnologia
Article - Open Access Idioma principal | Segundo principal

MÉTODO HÍBRIDO PARA DETECÇÃO E REMOÇÃO DE SOMBRAS EM IMAGENS

HYBRID METHOD FOR DETECTION AND REMOVAL OF IMAGE SHADOWS

Figueredo, Dr. Marcos Batista; Junior, Eugenio Rocha Silva

Article:

Sombras em uma imagem podem distorcer informações sobre a forma e orientação do objeto e até sobre a fonte de luz. Assim, a detecção e remoção de sombras é uma tarefa crucial e inevitável para alguns processos de visão computacional. Este artigo propõe uma estrutura simples usando o espaço de cores HSV para detectar e remover sombras de imagens. Inicialmente, uma abordagem probabilística da intensidade no espaço de cores HSV é proposta para a detecção de sombras. Depois que as sombras são identificadas, um modelo de densidade de sombra é aplicado A principal contribuição do modelo proposto é que, após a remoção de sombras, não há transição brusca entre as partes sombreadas e as partes não sombreadas, e todos os detalhes nas regiões sombreadas permanecem intactos.

Shadows in an image can distort information about the shape and orientation of the object and even about the light source. Thus, detecting and removing shadows is a crucial and inevitable task for some computer vision processes. This article proposes a simple structure using HSV color space to detect and remove shadows from images. Initially, a probabilistic approach to HSV color space intensity is proposed for shadow detection. Once shadows are identified, a shadow density model is applied. The main contribution of the proposed model is that after the removal of shadows, there is no sharp transition between the shaded parts and the unshaded parts, and all the details in the regions. Shaded shades remain intact.

Download (PDF)

Palavras-chave:

DOI: 10.5151/siintec2019-84

Referências bibliográficas
  • [1] 1 C.X. Jiang and M.O. Ward. Shadow segmentation and classification in a
  • [2] constrained environment. CVGIP: Image Understanding, 1994, 59(2):213 – 225.
  • [3] 2 S. H. Khan, M. Bennamoun, F. Sohel, and R. Togneri. Automatic shadow
  • [4] detection and removal from a single image. IEEE Transactions on Pattern
  • [5] Analysis and Machine Intelligence, 38(3):431–446, 2016.
  • [6] 3 N. Su, Y. Zhang, S. Tian, Y. Yan, and X. Miao. Shadow detection and removal for
  • [7] occluded object information recovery in urban high-resolution panchromatic
  • [8] satellite images. IEEE Journal of Selected Topics in Applied Earth Observations
  • [9] and Remote Sensing, 9(6):2568– 2582, 2016.
  • [10] 4 Fei Kou, Weihai Chen, Changyun Wen, and Zhengguo Li. Gradient domain
  • [11] guided image filtering. IEEE Transactions on Image Processing, 24(11):4528–5 Anat Levin, Assaf Zomet, Shmuel Peleg, and Yair Weiss. Seamless image
  • [12] stitching in the gradient domain. In European Conference on Computer Vision,
  • [13] pages 377–389. Springer, 2004.
  • [14] 6 Patrick Pérez, Michel Gangnet, and Andrew Blake. Poisson image editing. ACM
  • [15] Transactions on graphics (TOG), 22(3):313–318, 2003.
  • [16] 7 Graham D Finlayson, Steven D Hordley, Cheng Lu, and Mark S Drew. On the
  • [17] removal of shadows from images. IEEE transactions on pattern analysis and
  • [18] machine intelligence, 28 (1):59–68, 2005.
  • [19] 8 Hazewinkel, M, Poisson equation, Encyclopedia of Mathematics, Springer
  • [20] Science+Business Media B.V. / Kluwer Academic Publishers, Disponível em:
  • [21] <https://www.encyclopediaofmath.org/index.php/Poisson_equation>. Acesso em: 18
  • [22] ago. 2019.
  • [23] 9 T. F. Y. Vicente, M. Hoai, and D. Samaras. Leave-one-out kernel optimization for
  • [24] shadow detection and removal. IEEE Transactions on Pattern Analysis and
  • [25] Machine Intelligence, 40 (3):682–695, 2018.
  • [26] 10 M. Xu, J. Zhu, P. Lv, B. Zhou, M. F. Tappen, and R. Ji. Learning-based shadow
  • [27] recognition and removal from monochromatic natural images. IEEE Transactions
  • [28] on Image Processing, 26(12):5811–5824, 2017.
  • [29] 11 K. He, R. Zhen, J. Yan, and Y. Ge. Single-image shadow removal using 3D
  • [30] intensity surface modeling. IEEE Transactions on Image Processing, 26(12):6046–
  • [31] 6060, 2017.
  • [32] 12 L. Zhang, Q. Zhang, and C. Xiao. Shadow remover: Image shadow removal
  • [33] based on illumination recovering optimization. IEEE Transactions on Image
  • [34] Processing, 24(11):4623–4636, 2015.
  • [35] 15 Otsu, Nobuyuki. A threshold selection method from gray-level histograms. IEEE
  • [36] Trans. Sys., Man., Cyber. 9 (1): 62–66. 1979.
Como citar:

Figueredo, Dr. Marcos Batista; Junior, Eugenio Rocha Silva; "MÉTODO HÍBRIDO PARA DETECÇÃO E REMOÇÃO DE SOMBRAS EM IMAGENS", p-676-683. In: Anais do V Simpósio Internacional de Inovação e Tecnologia. São Paulo: Blucher, 2019.
ISSN 23577592, DOI 10.5151/siintec2019-84

últimos 30 dias

118
downloads

199
visualizações

692
indexações