I'm trying to find red cells on a green background. I have tried already a few methods (RGB, HSV...) but I'm struggling with very light red cells. The best method up to now is to take the RGB values and set the highest value to 255 and the rest to 0, e.g. if (167, 20, 63) -> (255, 0, 0). However this is risky and do not work very well with light cells. Can anyone help with some methods hints? I attached an example.
Edit: There is no need to find the cells, only a pixel number value is enough.
Any help is very appreciated!
Stefanie
Related
So I know basically nothing about colors apart from the very basics.
I have a color I'm trying to mimic.
I copied it, stuck it in paint, and used the color feature to get the RGB HSL numbers. Great!
RGB: 0; 49; 70
HSL: 132; 240; 33
The issue: When I try to manually input them into Excel, it "autocorrects" the RGB values after I enter in the HSL, and it "autocorrects" the HSL when I re-enter the RGB.
Why is this happening? Is this just an aspect to colors I know nothing about? Some limitation on Excel?
For reference, when I put in just the RGB, I'm much closer (but not quite there) on the color I'm looking for
HSL and RGB are two ways of "translating colors" into numbers.
HSL means Hue, Saturation, Lightness.
Hue is a degree on the color wheel from 0 to 360. 0 is red, 120 is green, 240 is blue.
Saturation is a percentage value; 0% means a shade of gray and 100% is the full color.
Lightness is also a percentage; 0% is black, 100% is white.
RGB means Red, Green, Blue, each of which is given a value between 0 and 255 in Excel.
Check this tool - https://www.w3schools.com/colors/colors_hsl.asp
If you put 0, 49, 70 for HSL you would see that it gets translated to 216, 141, 141 into RGB.
Excel is following the same logic, thus once you adjust the RGB the HSL gets automatically adjusted to represent the same color.
Excel colors are confusing because they don't follow the standard
Although the first reason you may have been confused was if you didn't know that RGB and HSL are two different ways of describing colors (and that every RGB color code has an equivalent HSL color code—see examples below), a second reason many people can get confused when selecting colors in Excel particularly is:
“Frustratingly, Excel does not handle HSL in the standard way. Instead, Excel measures all the numbers where 0 is the lowest and 255 is the biggest. But, it’s a quirk we can handle.” - https://exceloffthegrid.com/convert-color-codes/
“This approach assumes that each of your HSL values can be express in the range of 0 to 255. If, however, your HSL values are either an angle (for hue) or a percentage (for saturation and luminance), then you'll need to convert them manually before entering them in step 6. You can convert an angle value by multiplying the angle by 255 and then dividing by 360. Percentages can be converted by multiplying them by 2.55.” - https://excelribbon.tips.net/T013535_Converting_HSL_to_RGB.html
“To change the lightness (adding white) or darkness (adding black), drag your selection up and down the luminance scale on the right. Notice that the Lum value increases as the color gets lighter. Full luminance is 255 (white), and setting Lum to 0 results in black regardless of the hue and saturation settings.” - https://support.microsoft.com/en-us/office/choosing-colors-in-the-colors-dialog-box-c3d59ddf-65a7-4e62-aad7-f7b8d7684a2d
Examples of converting RGB color codes to HSL
rgb(0, 49, 70) = hsl(198, 100%, 14%)
These independent sites agree with Google that that RGB code converts to that HSL code:
https://colorpicker.me/#003146
https://hslpicker.com/#003146
So if someone told you that rgb(0, 49, 70) was equivalent to hsl(132, 240, 33), they were mistaken (even when using Excel's non-standard way of calculating HSL).
My question is about color tracking... What is the color threshold value for white in python ? I need to track the white color alone in a group of pictures. So I need to separate the white color. In order to do so I need to know the threshold value of white color...
It depends on your pictures. Assuming that you're going to threshold using the RGB values, the RGB value for white is (255, 255, 255). But this value holds true for pure white color. If you have real-world pictures, you might have clearly white color at certain areas in your image but they wouldn't have the value (225, 255, 255). Factors like the shadow, lighting conditions, angle etc. contribute to the variance from pure white color value.
In order to threshold correctly, you need to check the range of values for your set of pictures. I recently worked on a similar problem and for my case, the range of values was as follows:
Red channel: 200-255
Green channel: 180-255
Blue channel: 140-255
But please note that this accepts a lot of variation of white like light yellow. It will highly depend on your case so make sure you check the range on your data.
One way of that can be by displaying/showing your image using skimage and then hovering over white areas, it will display the RGB value on the bottom right corner of the image. Here is the code for showing an image in skimage:
from skimage import io
def show(img):
io.imshow(img)
io.show()
You can create a range of values/threshold from the values you notice this way.
I have compiled a heat map with the following code:
compounds=uiimport('matrix.txt');
names=uiimport('names.txt');
stages=['Stage ' '3';'Stage ' '6';'Stage ' '9';];
imagesc(compounds.matrix);
colorbar;
colormap('winter');
title('Heat map of ...')
ylabel('Compounds');
xlabel('Developmental stage');
set(gca,'YTick',1:21,'YTickLabel',names.names)
set(gca,'XTick',1:3,'XTickLabel',stages)
Everything works fine and i do not want to change the code as it comes from uplading my data, however i cannot figure out how to add a code (or should i say what kind of code) that would adjust my heatmap colors at a particular scale numbers.
What i would like, is for it to be red to green (0 to 10), with a value 1 as black.
Additionally, my scale is 0 to 100, but i would like the scale to be visible from 1 to 10 and then have a gap and see just 100 at the top.
Many thanks in advance.
There isn't a language requirement listed here, but can I assume you are using matplotlib or matlab.
You can specify the first part of your request by doing breaks. Setting certain ranges. They have a good explanation here
Color a heatmap in Python/Matplotlib according to requirement
A more technical method is to actually create your own colormap
colors = [('white')] + [(cm.jet(i)) for i in xrange(1,256)]
new_map = matplotlib.colors.LinearSegmentedColormap.from_list('new_map', colors, N=256)
Here you can specify color for particular ranges in this example white is 0 and the rest is the standard flowing range
Credit:
Make reverse diagonals white in heatmap
Just a short explanation how I came to this question. I have a ruby module which is basically a hash that gives me HTML "colours", like "slateblue", and gives me back an Array that holds the R,G,B values, like [106, 90, 205] for slateblue.
I googled how to make these R,G,B values into a lighter colour (for mouse cursor on hover effect), and several people told other people when they had a similar problem to just increase the R,G,B values. My current solution, which is a hack, is to add to the R,G,B values, like +20 (capped at 255), and then convert this into a hexstring #FF0000 something.
This seems to work okish but here is the thing now - I have absolutely no understanding about why this works.
Is it so that the 0 always denotes the lowest value of R/G/B and 255 the highest? If so, why is it capped at 255 and not at, don't know, 1024 or some other arbitrary number?
Using 8-bits per color channel - one each for red, green, and blue - yields a large number of colors (2^24 or 16777216), and is sufficient to be used in most applications. Note that there are other color formats with higher precision though.
0 is used for black, while 255 (the maximum stored in 8-bits) denotes "full-on" color.
Adding a specific number to each channel moves the entire color toward (255, 255, 255), or White. If you would like to be more exact in your lightening of the color, you might try converting your RGB color to HSL, doing your addition to the light component only, then converting back to RGB.
You can start research of HSL here: http://en.wikipedia.org/wiki/HSL_and_HSV
I have more then 1 week reading about selective color change of an image. It meand selcting a color from a color picker and then select a part of image in which I want to change the color and apply the changing of color form original color to color of color picker.
E.g. if I select a blue color in color picker and I also select a red part in the image I should be able to change red color to blue color in all the image.
Another example. If I have an image with red apples and oranges and if I select an apple on the image and a blue color in the color picket, then all apples should be changing the color from red to blue.
I have some ideas but of course I need something more concrete on how to do this
Thank you for reading
As a starting point, consider clustering the colors of your image. If you don't know how many clusters you want, then you will need methods to determine whether to merge or not two given clusters. For the moment, let us suppose that we know that number. For example, given the following image at left, I mapped its colors to 3 clusters, which have the mean colors as shown in the middle, and representing each cluster by its mean color gives the figure at right.
With the output at right, now what you need is a method to replace colors. Suppose the user clicks (a single point) somewhere in your image, then you know the positions in the original image that you will need to modify. For the next image, the user (me) clicked on a point that is contained by the "orange" cluster. Then he clicked on some blue hue. From that, you make a mask representing the points in the "orange" cluster and play with that. I considered a simple gaussian filter followed by a flat dilation 3x5. Then you replace the hues in the original image according to the produced mask (after the low pass filtering, the values on it are also considered as a alpha value for compositing the images).
Not perfect at all, but you could have a better clustering than me and also a much-less-primitive color replacement method. I intentionally skipped the details about clustering method, color space, and others, because I used only basic k-means on RGB without any pre-processing of the input. So you can consider the results above as a baseline for anything else you can do.
Given the image, a selected color, and a target new color - you can't do much that isn't ugly. You also need a range, some amount of variation in color, so you can say one pixel's color is "close enough" while another is clearly "different".
First step of processing: You create a mask image, which is grayscale and varying from 0.0 to 1.0 (or from zero to some maximum value we'll treat as 1.0), and the same size as the input image. For each input pixel, test if its color is sufficiently near the selected color. If it's "the same" or "close enough" put 1.0 in the mask. If it's different, put 0.0. If is sorta borderline, put an in-between value. Exactly how to do this depends on the details of the image.
This might work best in LAB space, and testing for sameness according to the angle of the A,B coordinates relative to their origin.
Once you have the mask, put it aside. Now color-transform the whole image. This might be best done in HSV space. Don't touch the V channel. Add a constant to S, modulo 360deg (or mod 256, if S is stored as bytes) and multiply S by a constant chosen so that the coordinates in HSV corresponding to the selected color is moved to the HSV coordinates for the target color. Convert the transformed S and H, with the unchanged L, back to RGB.
Finally, use the mask to blend the original image with the color-transformed one. Apply this to each channel - red, green, blue:
output = (1-mask)*original + mask*transformed
If you're doing it all in byte arrays, 0 is 0.0 and 255 is 1.0, and be careful of overflow and signed/unsigned problems.