The color-grading from steveseguin

color-grading's People

Contributors

Stargazers

Watchers

color-grading's Issues

I made something wrong, but not sure where.

Fantastic project to learn calibration @steveseguin, thank you!

I have a color reference card with known RGB values so I need exactly this project to make a correction LUT in OBS.
I shot a full size ref card with some slight perspective and it also have numbers. Maybe because of this, it doesn't see the first colors and I get more RGB values in histogram and computed LUT is incorrect .

Is there a way, in Step 4, to see/tweak which point/area it "sees" (uses)?
Here are my files enclosed to get an idea.

expected:

ref card

Tweak the code to deal with target arrays with sizes other than 3x3

I tweaked the code to automatically deal with targets that aren't 3x3 grids:

Suggested step four:

color = ('r','g','b')
for x in range(total_rows):
	for y in range(total_columns):
		for i,col in enumerate(color):
			seg = img3[int(x*w/total_rows):int((x+1)*w/total_rows),int(y*h/total_columns):int((y+1)*h/total_columns),:] # We assume each color block is the same size and that the image is cropped correctly
			histr = cv2.calcHist([seg],[i],None,[256],[0,256]) # Create a histogram
			j = x*total_columns+y+1
			wrong[j-1][i]=np.where(np.max(histr)==histr)[0][0] # What is the PEAK color for each block? use this rather than the average RGB value.
			plt.subplot(total_rows,total_columns,j)
			plt.plot(histr,color = col) # Plot, so you can se what's going on.
			plt.xlim([0,256])
			
 # If successful, you should see some plots below. Each plot should have 1 blue, 1 red, and 1 green peak each. 
test = np.array(wrong).reshape((total_rows,total_columns,3)).astype("uint8") 
test = cv2.cvtColor(test, cv2.COLOR_RGB2BGR)
test = cv2.resize(test,(200,150),interpolation=cv2.INTER_NEAREST)
cv2_imshow(test)  ## THIS IS WHAT THE INPUT IT IS; confirm it matches the photo you uploaded.

Recommend Projects