Depends on how big the area is. More precisely, what is the maximum distance from the set of missing pixels to the nearest valid pixel. The general technique is called Infilling. Here's just an example from Google search result. http://research.microsoft.com/pubs/67276/criminisi_tip2004.pdf When the text is very thin, a masked local averaging will do the job. img = imread( filename ) ; [ nrows, ncols, nchan ] = size ( img ) ; imgColor = double ( img ) .* (1 / 255) ; imgGray = rgb2gray ( imgColor ) ; % Since the text is at the bottom of the image, % we specify this condition to prevent mis-classification % of pixels from the lightly-colored sky. rowLowerIndex = round ( (nrows - 1) * 0.8 ) + 1; maskLower = false ( nrows, ncols ) ; maskLower ( rowLowerIndex : end , : ) = true ; % extracts the pixel locations of the text maskText = logical ( imgGray >= 0.9 ) & maskLower ; % invert the mask, because we will use non-text pixels % to cover up text pixels. maskNonText = ~ maskText ; % Convert the mask into a zero-one matrix % (floating point), which can be used as an % element-wise weighting matrix floatMaskNonText = repmat( double ( maskNonText ) , [ 1, 1, 3 ] ); % ... 省略,您懂的 imgNumerator = imfilter ( imgColor .* floatMaskNonText , fspecial ( ... ) ); imgDenominator = imfilter ( floatMaskNonText, fspecial ( ... ) ) ; % then element-wise divide.
w*3
26 楼
应该file separate 吧,还有什么不对? 我好奇LZ申请到ITIN没有 你也要开始搞报税的了吗
【在 A**l 的大作中提到】 : lz很多概念都是不对得。。。你没看我前面的分析?
r*y
27 楼
用胸腔,嗓子不用力,增加音量不变音高。
【在 s**l 的大作中提到】 : 经常对口型?
p*e
28 楼
欢迎MM常来玩,呵呵
【在 m***5 的大作中提到】 : 好久没来,好像出了很多事,MM到这里做版主啦,恭喜恭喜
j*5
29 楼
买个手机装Garmin XT加国内地图吧
w*s
30 楼
我察,兄弟你太牛B了, 送福利。
pixels
【在 k**********g 的大作中提到】 : Depends on how big the area is. : More precisely, what is the maximum distance from the set of missing pixels : to the nearest valid pixel. : The general technique is called Infilling. : Here's just an example from Google search result. http://research.microsoft.com/pubs/67276/criminisi_tip2004.pdf : When the text is very thin, a masked local averaging will do the job. : img = imread( filename ) ; : [ nrows, ncols, nchan ] = size ( img ) ; : imgColor = double ( img ) .* (1 / 255) ; : imgGray = rgb2gray ( imgColor ) ;
a*4
31 楼
F1是没有办法file jointly的,其实LZ如果不等到现在,在你LP转F1之前file倒是有可 能可以first year choice的
en,sometimes forget is more important than remember, easier than forgive... just like xuxu is more important than drink that's why watering is vely vely important...
【在 e*******s 的大作中提到】 : It's a good way to help you to forget...
【在 k**********g 的大作中提到】 : Depends on how big the area is. : More precisely, what is the maximum distance from the set of missing pixels : to the nearest valid pixel. : The general technique is called Infilling. : Here's just an example from Google search result. http://research.microsoft.com/pubs/67276/criminisi_tip2004.pdf : When the text is very thin, a masked local averaging will do the job. : img = imread( filename ) ; : [ nrows, ncols, nchan ] = size ( img ) ; : imgColor = double ( img ) .* (1 / 255) ; : imgGray = rgb2gray ( imgColor ) ;