-
公开(公告)号:US07102638B2
公开(公告)日:2006-09-05
申请号:US10392000
申请日:2003-03-19
申请人: Ramesh Raskar , Jingyi Yu , Adrian Ilie
发明人: Ramesh Raskar , Jingyi Yu , Adrian Ilie
IPC分类号: G09G5/02
摘要: A method generates an image with de-emphasized textures. Each pixel in the image is classified as either a silhouette edge pixel, a texture edge pixels, or a featureless pixel. A mask image M(x, y) is generated, wherein an intensity of a given pixel (x, y) in the mask image M(x, y) is zero if the pixel (x, y) is classified as the texture edge pixel, is d(x, y) if the pixel (x, y) is classified as the featureless pixel, and is one if the pixel (x, y) is classified as the silhouette edge pixel. An intensity gradient ∇I(x, y) is determined in the masked image, and the intensity gradients in the masked image are integrated according to G(x, y)=∇I(x, y). M(x, y). Then, an output image I′ is generated by minimizing |∇I′−G|, and normalizing the intensities in the output image I′.
摘要翻译: 方法生成具有去加重纹理的图像。 图像中的每个像素被分类为轮廓边缘像素,纹理边缘像素或无特征像素。 生成掩模图像M(x,y),其中如果像素(x,y)被分类为纹理边缘,则掩模图像M(x,y)中的给定像素(x,y)的强度为零 像素,如果像素(x,y)被分类为无特征像素,则为d(x,y),并且如果像素(x,y)被分类为轮廓边缘像素则为像素(x,y)。 在掩蔽图像中确定强度梯度∇I(x,y),并根据G(x,y)=∇I(x,y)将屏蔽图像中的强度梯度积分。 M(x,y)。 然后,通过最小化|∇I'-G |并且使输出图像I'的强度标准化来生成输出图像I'。
-
公开(公告)号:US07103227B2
公开(公告)日:2006-09-05
申请号:US10392429
申请日:2003-03-19
申请人: Ramesh Raskar , Adrian Ilie , Jingi Yu
发明人: Ramesh Raskar , Adrian Ilie , Jingi Yu
CPC分类号: G06T5/20 , G06T5/002 , G06T5/003 , G06T5/50 , G06T7/12 , G06T7/174 , G06T11/00 , G06T15/02 , G06T2207/20192
摘要: A method enhances images of a naturally illuminated scene. First, a set of images Ii(x, y) is acquired of a scene. Each image is acquired under a different uncontrolled illumination. For each image Ii(x, y), intensity gradients ∇Ii(x, y) are determined, and each image is weighted with weights according to the intensity gradients. The weighted images are then combined to construct an enhanced image I′.
摘要翻译: 一种方法可以增强自然照明场景的图像。 首先,获取场景的一组图像I SUB>(x,y)。 每个图像在不同的不受控照明下被采集。 对于每个图像,我确定(x,y),强度梯度ξI(x,y),并且根据强度梯度对每个图像进行权重加权 。 然后将加权图像组合以构建增强图像I'。
-
公开(公告)号:US20070098288A1
公开(公告)日:2007-05-03
申请号:US10392026
申请日:2003-03-19
申请人: Ramesh Raskar , Adrian Ilie
发明人: Ramesh Raskar , Adrian Ilie
IPC分类号: G06K9/40
CPC分类号: H04N1/3871 , G06T5/002 , G06T5/20 , G06T7/12 , G06T2207/10016 , G06T2207/20028 , G06T2207/20192 , H04N1/58 , H04N1/6086
摘要: A method enhances a low quality video. An input video Vjin is acquired of a scene under uncontrolled illumination. The input video Vjin is smoothed to produce a noise-reduced video Sj. Temporal gradients Gtj (x, y) are determined for each frame j of the noise reduced video Sj. The gradients Gtj (x, y) are thresholded to generate a binary mask Mj(x, y). For each frame, weights are determined according Wj (x, y)=F(Mk(x, y), and a spatial gradient field is determined according to Gxyj (x, y). Then, an output video Vjout is constructed from the spatial gradient field Gxyj.
摘要翻译: 一种方法增强了低质量的视频。 在不受控照明下的场景中获取 SUB>中的输入视频V SUP> 。 平滑了 SUB>中的输入视频V SUB,以产生噪声降低视频S SUP。 对于降噪视频S SUP的每个帧j确定时间梯度G(x,y)。 阈值化梯度G(x,y)以产生二进制掩码M(x,y)。 对于每个帧,根据W(x,y)= F(M,k)(x,y)确定权重,并且根据G确定空间梯度场 (x,y),然后,从空间梯度构成输出视频V OUT, 字段G SUB> SUP> j SUP>。
-
-