Let's say I have the following image:
When I use the ImageMagick, the command identify -format %k image.png counts the number of different colors on this image. My output for this image is 1031, which would mean that this image has 1031 different colors. However, visually saying I have around 5 different colors:
black
white
grey
light green
dark green
I'd like to find a way of getting a list of the number of major color pixels. For example, let's say that I have 1000 pixels on the last image, so I'd expect something like:
black:10
white:50
grey:200
light green:690
dark green:50
Is there any way to make ImageMagick display output values like the last ones? Just so I'll know the percentage of the main colors on my image?
You can load the image, tell ImageMagick you want it reduced to 5 colours and then get the histogram like this:
magick 0N9iv.png -colors 5 -verbose info: | grep -A5 Histogram:
Histogram:
459: (181.83,198.952,162.503,229.414) #B6C7A3E5 srgba(71.306%,78.0204%,63.7265%,0.899662)
26749: (182.035,228.807,158.076,254.874) #B6E59EFF srgba(71.3861%,89.7283%,61.9907%,0.999506)
577: (191.945,227.688,174.459,100.095) #C0E4AE64 srgba(75.2725%,89.2894%,68.4152%,0.392528)
727: (196.666,220.169,180.959,255) #C5DCB5FF srgba(77.124%,86.3407%,70.9642%,1)
4384: (219.415,217.829,213.189,255) #DBDAD5FF srgba(86.045%,85.423%,83.6034%,1)
Note that means that ImageMagick chooses the colours itself. If you specifically want your image mapped to your own particular set of colours, make a palette of those colours like this:
magick xc:black xc:white xc:gray xc:lightgreen xc:darkgreen +append palette.png
And then tell ImageMagick to map all the colours to those 5 colours and see the output:
magick 0N9iv.png +dither -remap palette.png -verbose info: | grep -A5 Histogram:
Sample output
Histogram:
81: (0,0,0,255) #000000FF black
22: (126,126,126,255) #7E7E7EFF gray
28270: (144,238,144,255) #90EE90FF LightGreen
4523: (255,255,255,255) #FFFFFFFF white
Colormap entries: 5
You can also select just the histogram like this:
magick 0N9iv.png +dither -remap palette.png -format %c histogram:info:-
Sample Output
81: (0,0,0,255) #000000FF black
22: (126,126,126,255) #7E7E7EFF gray
28270: (144,238,144,255) #90EE90FF LightGreen
4523: (255,255,255,255) #FFFFFFFF white
Related
Friends,
I have a stack of color-scanned images. Some are from regular white paper with text or images, others were scanned from colored paper (blank pages, same green colored paper used.)
I'd like to identify these colored paper images. Problems:
paper's color ("background") is not scanned very uniformly, often has a wavy or structured pattern
green tone is quite different depending on the scanner used
scanner does not catch the full sheet resulting in a white or shadowed "border" around green area
My idea was to see if say 90% of the image is some sort of green and tried using a sorted histogram. But because of (1) and esp. (2) I have a hard time picking a working color value from the histogram data.
Any help appreciated!
Edit:
Here are three sample images, scanned from the same sheet of paper.
Have a look at HSV colourspace on Wikipedia - specifically this diagram.
It should be a better place to find the colour of your images, regardless of scanner and calibration.
Now, let's create a lime-green, yellow and cyan block and derive its colour using ImageMagick:
magick -size 100x100 xc:lime -colorspace HSV -channel 0 -separate -format "%[fx:mean*360]" info:
120
magick -size 100x100 xc:yellow -colorspace HSV -channel 0 -separate -format "%[fx:mean*360]" info:
60
magick -size 100x100 xc:magenta -colorspace HSV -channel 0 -separate -format "%[fx:mean*360]" info:
300
magick -size 100x100 xc:cyan -colorspace HSV -channel 0 -separate -format "%[fx:mean*360]" info:
180
Hopefully you can see we are correctly calculating the Hue angle. Now to your image. I have added an artificial frame so you can see how to remove the edges:
We can remove the frame like this:
magick YOURSCAN.jpg -gravity center -crop 80% cropped.jpg
So, my complete suggestion would be to crop and convert to HSV and check the mean Hue. You could also test if the image is fairly saturated so it doesn't pick up grey-ish, uncoloured images. You could also test the variance in the Hue channel to see if there are many different colours - or the spread of the hues is large and reject ones where it is large.
magick YOURSCAN.jpg -gravity center -crop 80% -colorspace HSV -channel 0 -separate -format "%[fx:mean*360]" info:
Just for reference, your 3 images come up with the following Hue angles on a scale of 0..360:
79, 68, 73
I would suggest you test a few more samples to establish a reasonable range.
I would like to create a hotfolder for my motion camera, into which I can drop images marking areas, which should be excluded in motion recognition via a *pgm mask. On these images, there is a small area marked with a transparent box with a magenta colored outline. My aim is to replace this box and outline with a black solid box and the rest of the image with white. (Tried to post samples here, but not enough reputation to do so.)
I know how to do this "by foot" using gimp, but I cannot figure out a clever and simple way achieving this with imagemagick.
I tried googling for solutions with -trim and -virtual-pixel, but no luck. Any help would be appreciated.
I'll do this step-by-step so you can see the intermediate parts in case you are on Windows and bash doesn't work.
First, let's make make everything that is not within 10% of your magenta colour, namely rgb(225,75,130), into lime green:
magick source.jpg -fill lime -fuzz 10% +opaque "rgb(225,75,130)" result.png
Ok, now let's get the trim box - i.e. all the constant junk that ImageMagick could trim off to focus on the magenta bit.
magick source.jpg -fill black -fuzz 10% +opaque "rgb(225,75,130)" -format '%#' info:
14x66+426+118
So your magenta box is 14x66pixels and located at offset 426,118 from the top-left. Now we want to get those in bash variables w,h,x,y. We need to change x and + into spaces using tr:
read w h x y < <(magick source.jpg -fill black -fuzz 10% +opaque "rgb(225,75,130)" -format '%#' info: | tr 'x+' ' ')
If we print this we get:
echo $w, $h, $x, $y
14, 66, 426, 118
Now we want to draw a rectangle, but that needs top-left and bottom-right, so we need to do some maths:
((x1=x+w))
((y1=y+h))
Ok, now we can load the original image, make it fully white, then draw our black rectangle:
magick source.jpg -threshold -1 -fill black -draw "rectangle $x,$y $x1,$y1" -depth 8 mask.pgm
So, the whole thing boils down to:
#!/bin/bash
read w h x y < <(magick source.jpg -fill black -fuzz 10% +opaque "rgb(225,75,130)" -format '%#' info: | tr 'x+' ' ')
echo $w, $h, $x, $y
((x1=x+w))
((y1=y+h))
magick source.jpg -threshold -1 -fill black -draw "rectangle $x,$y $x1,$y1" -depth 8 mask.pgm
There are other (maybe more elegant) ways of doing it, using flood-fills and/or connected components but I didn't want it to rely on your magenta box being "watertight", i.e. not rely on the sides being continuous and complete.
Also, if the size of your images is known and constant, you can avoid reloading the original and making it white by thresholding like I do in the last line and just create a canvas of the known dimensions, i.e.:
magick -size ${W}x${H} xc:white -fill black -draw "rectangle $x,$y $x1,$y1" -depth 8 mask.pgm
I would like to ignore/filter white and those that are close to white when using this type of command
convert $file -colors 10 -format "%c" histogram:info:
I have some images that are predominantly white in their content and i'd like to focus on the other colours.
I thought i could use something like -fuzz 20% -fill "#0000ff" -opaque white to change all white and white'ish into a simple blue which i could then grep out in the histogram output but no luck.
Can anyone point me in the correct direction or offer an example please ?
Let's say for example i have a photo that contains a very overexposed sky i.e lot's of white. I'd like the histogram output to ignore that part of the photo and instead output the other top 10 colours that are in the photo.
Thanks
fLo
Currently, I have thousands of pictures where I want to get rid of white space around the object, but not inside the object. However, the white space is not constant. See pictures below.
Picture 1 :
Picture 2 :
Picture 3 :
Options like this don't work as they delete all white space in the images
convert image.png -fuzz 10% -transparent white output.png
Any suggestions?
What do you mean by getting rid of the white space? If you mean make it transparent, then in ImageMagick 6, you can do a whole folder of images using the mogrify command as follows:
The process involves padding the image with 1 pixel border of white to ensure white is all around. Then doing a fuzzy flood fill. Then remove the extra one pixel all around. I have assumed that the background color is nearly white. Change the fuzz value as desired.
Create a new directory to hold the output (or backup your input directory)
Change directory to the one holding your input images
mogrify -path path_to/outdirectory -format png -bordercolor white -border 1 -fuzz 20% -fill none -draw "matte 0,0 floodfill" -shave 1x1 *.png
Here are the resulting images.
Of course you will have to download them to see that they are transparent.
Note that the images that I downloaded were JPG. So if your images are jpg and not png as in your command, then change *.png to *.jpg. Or just use * to change all formats.
See
https://imagemagick.org/Usage/basics/#mogrify
https://imagemagick.org/Usage/draw/#matte
If using ImageMagick 7, change matte to alpha and change mogrify to magick mogrify.
I have an image in .jpg format with white background color. I want to remove the white background color to transparent in Imagemagick. I tried many ways but still the white background can not be removed. Can some one help me to solve this.
You cannot have transparent background colors in your JPEGs. The JPEG file format doesn't support transparency.
If you need transparent background, you need to convert the JPEG to
either PNG (high quality, filesize possibly larger than JPEG)
or GIF (in case you can tolerate low quality and a range of maximally 255 colors).
Example command:
convert your.jpg -transparent white your.png
First, you need to convert the image format from .jpg to .png format, because JPEG does not support transparency. Then use this command:
convert image1.png -fuzz 20% -transparent white result.png
The -fuzz option allows the specified percentage deviation from the pure white colour to be converted to transparent as well. This is useful, for example, when your image contains noise or subtle gradients.
I just found a very neat thing!
magicwand 1,1 -t 20 -f image -r outside -m overlay -o 0 image.jpg imgOutput.png
It is a Fred Weinhaus bash script that can be downloaded from here (for non commercial use only). Also there has about 250 scripts!! and this one is amazing! it did exactly the trick, to remove all background while keeping the inner image dots untouched!
At his page, there are several images as examples so you pick what you need to put on the command line!
The initial position 1,1 is a general guesser saying all the contour is background.
Pay attention that the output must be ".png"
This is my solution without magicwand (replace magick by convert for im < 7.0):
magick img.png -fuzz 20% -fill none -draw "alpha 1x1 floodfill" result.png
Get the background automatically and remove it :
bg=$(convert input.png -format "%[pixel:p{0,0}]" info:)
convert input.png -fuzz 20% -transparent "$bg" output.png