Automatically remove the background from the photographed text in Photoshop or ImageMagick
I want to share a trick that helps a little to improve the quality when printing in b / w of all kinds of abstracts, if for some reason they are “digitized” with a camera instead of a scanner.
Actually, the main problem with such photos, unlike scans, is the unevenness of the background brightness, sometimes so much so that the text on such a dark background becomes poorly readable after printing on a laser b / w printer. As an example of such "unevenness":
The idea is to select something similar to its background from the page and “subtract” it from it itself, and thereby destroy at least some of the background and brightness differences.
Get to the point. Suppose we have a couple of dozen, photographed in similar conditions pages. Like this, but worse:
')
Open any of the pages in Photoshop and make a copy of the main layer:
Invert the top layer:
blur it with a gaussian blur
... it is very important to choose the blur radius correctly: on the one hand it should be larger than the characteristic size of the letters - we need to destroy all the details, on the other hand too blurred layer cannot convey the gradient of the background that we are trying to lime. The optimal radius is somewhere around 1.5–3 the size of the row height:
it only remains to change the layer blending method, the best thing is what Color Dodge does:
all gradient is no more:
You can merge the layers together and twist the brightness and levels to taste (everything depends heavily on the specific conditions of the shooting, while the previous steps are highly universal).
Now, after practicing, you can proceed to automate the process:
Cancel everything done with the picture and start recording the macro:
under the record we repeat everything that we did at the beginning of the article - from the cloning of the layer, to the layer of layers and brightness twist.
Stop the macro recording and go to File => Automate => Batch ...
There we choose where to get pictures of the pages and where to save them later:
The process is not very fast, but fully automated.
I hope someone will come in handy, most not relevant for a long time.
UPD: I recommend after passing through the layers to pass Unsharp Mask (I used the parameters: radius 5 pixels / 150%) (c) not_ice
UPD: you can do the same with ImageMagick , for example, with this command
Criticism and improvements are welcome.
Actually, the main problem with such photos, unlike scans, is the unevenness of the background brightness, sometimes so much so that the text on such a dark background becomes poorly readable after printing on a laser b / w printer. As an example of such "unevenness":
The idea is to select something similar to its background from the page and “subtract” it from it itself, and thereby destroy at least some of the background and brightness differences.
Get to the point. Suppose we have a couple of dozen, photographed in similar conditions pages. Like this, but worse:
')
Open any of the pages in Photoshop and make a copy of the main layer:
Invert the top layer:
blur it with a gaussian blur
... it is very important to choose the blur radius correctly: on the one hand it should be larger than the characteristic size of the letters - we need to destroy all the details, on the other hand too blurred layer cannot convey the gradient of the background that we are trying to lime. The optimal radius is somewhere around 1.5–3 the size of the row height:
it only remains to change the layer blending method, the best thing is what Color Dodge does:
all gradient is no more:
You can merge the layers together and twist the brightness and levels to taste (everything depends heavily on the specific conditions of the shooting, while the previous steps are highly universal).
Now, after practicing, you can proceed to automate the process:
Cancel everything done with the picture and start recording the macro:
under the record we repeat everything that we did at the beginning of the article - from the cloning of the layer, to the layer of layers and brightness twist.
Stop the macro recording and go to File => Automate => Batch ...
There we choose where to get pictures of the pages and where to save them later:
The process is not very fast, but fully automated.
I hope someone will come in handy, most not relevant for a long time.
UPD: I recommend after passing through the layers to pass Unsharp Mask (I used the parameters: radius 5 pixels / 150%) (c) not_ice
UPD: you can do the same with ImageMagick , for example, with this command
convert.exe src.jpg ( +clone -negate -virtual-pixel edge -gaussian-blur 30x10 ) -background none -compose colordodge -flatten result.jpg
Criticism and improvements are welcome.
Source: https://habr.com/ru/post/67976/
All Articles