Bokeh Display Based on Depth Information Extraction of Multi-focus Images
-
摘要: 针对普通相机无法拍摄出符合视觉美感的背景虚化图像问题, 提出一种基于多聚焦图像深度信息提取的背景虚化算法. 首先, 根据图像的光学成像原理,对多聚焦图像进行热扩散方程的数学建模, 对图像中的每个点自适应的初始化一个深度值; 其次, 根据图像每个点的热扩散系数对初始图像进行前向扩散, 用数值方法来解此偏微分方程, 通过正则化得到修正后的图像深度信息; 最后, 对图像深度信息进行分层, 利用多尺度高斯滤波对不同深度层的图像进行虚化处理. 本算法能够有效地提取出图像的深度信息, 并且获得的背景虚化效果也能够较好地符合光学原理.Abstract: It is difficult to take photos to meet the visual beauty bokeh effect by non single-lens reflex cameras. This paper presents a bokeh algorithm based on multi-focus image depth information extraction. Firstly, the mathematical model of heat conduction equation is built for the multi-focus images by optical imaging principle. Each point in the image is initiated by an adaptive depth value. Secondly, the forward diffusion is performed on the original image according to the thermal diffusion coefficient of every point in the image. The numerical method is used to solve the partial differential equation. The adjusted depth information of the image is obtained by regularization. Finally, the depth information of the image is layered. The images of different depth layers are blured by a multi-scale Gaussian filter. The algorithm proposed can effectively extract the image depth information and the bokeh effect is also able to meet the optical principle efficiently.
-
Key words:
- Multi-focus /
- depth extraction /
- diffusion /
- bokeh
-
[1] Subbarao M, Surya G. Depth from defocus: a spatial domain approach. International Journal of Computer Vision, 1994, 13(2): 271-294 [2] [2] Favaro P, Mennucci A, Soatto S. Observing shape from defocused images. International Journal of Computer Vision, 2003, 52(1): 25-43 [3] [3] Favaro P, Soatto S. A geometric approach to shape from defocus. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(2): 406-417 [4] [4] Favaro P, Soatto S, Burger M, Osher S J. Shape from defocus via diffusion. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2008, 30(2): 518-531 [5] [5] Zhuo S J, Sim T. Defocus map estimation from a single image. Pattern Recognition, 2011, 44(9): 1852-1858 [6] [6] Tang C, Hou C, Song Z. Defocus map estimation from a single image via spectrum contrast. Optics Letters, 2013, 38(10): 1706-1708 [7] [7] McIntosh L, Riecke B E, DiPaola S. Efficiently simulating the bokeh of polygonal apertures in a post-process depth of field shader. Computer Graphics Forum, 2012, 31(6): 1810- 1822 [8] [8] Kriener F, Binder T, Wille M. Accelerating defocus blur magnification. In: Proceedings of the 2013 SPIE, Volume 8667, Multimedia Content and Mobile Devices. Burlingame, California, USA, 2013. 86671Q-11 [9] [9] Wu J Z, Zheng C W, Hu X H, Xu F J. Rendering realistic spectral bokeh due to lens stops and aberrations. The Visual Computer, 2013, 29(1): 41-52 [10] Xiao Jin-Sheng, Shan Shan-Shan, Duan Peng-Fei, Tu Chao-Ping, Yi Ben-Shun. A fast image enhancement algorithm based on fusion of different color spaces. Acta Automatica Sinica, 2014, 40(4): 697-705(肖进胜, 单姗姗, 段鹏飞, 涂超平, 易本顺. 基于不同色彩空间融合的快速图像增强算法. 自动化学报, 2014, 40(4): 697-705) [11] An Yao-Zu, Lu Yao, Zhao Hong. An adaptive-regularized image super-resolution. Acta Automatica Sinica, 2012, 38(4): 601-608(安耀祖, 陆耀, 赵红. 一种自适应正则化的图像超分辨率算法. 自动化学报, 2012, 38(4): 601-608) [12] Zhang Y Q, Ding Y, Xiao J S, Liu J, Guo Z. Visibility enhancement using an image filtering approach. EURASIP Journal on Advances in Signal Processing, 2012, 2012(1): 1 -6 [13] Xiao J S, Sun L L. An iterative algorithm for maximal monotone multivalued operator equations. Acta Mathematica Scientia, 2001, 21(B2): 152-158 [14] Huang Zhi-Yin, Zhang Mao-Jun, Zhang Xin, Wang Wei. Multi-focus image fusion algorithm based on relative activity level. Microcomputer Information, 2009, 25(18): 289- 291(黄志银, 张茂军, 张鑫, 王炜. 基于相对清晰度的多焦距图像融合算法. 微计算机信息, 2009, 25(18): 289-291) [15] Xue Qian, Yang Cheng-Yi, Wang Hua-Xiang. Alternating direction method for salt-and-pepper denoising. Acta Automatica Sinica, 2013, 39(12): 2071-2076(薛倩, 杨程屹, 王化祥. 去除椒盐噪声的交替方向法. 自动化学报, 2013, 39(12): 2071-2076) [16] Chai Y, Li H F, Guo M Y. Multifocus image fusion scheme based on features of multiscale products and PCNN in lifting stationary wavelet domain. Optics Communications, 2011, 284(5): 1146-1158
点击查看大图
计量
- 文章访问数: 2103
- HTML全文浏览量: 106
- PDF下载量: 1617
- 被引次数: 0