从像素中获取 RGB 值

Question

I am working on a problem in which I am plotting a GeoTIFF image using matplotlib. I want to implement a function such that when I click on the image, the function must return the RGB value of the pixel on which I clicked, using matplotlib only.

I had seen all the previous solutions provided on Stack Overflow, none of those worked for me.

Here is my code:

src = rasterio.open("rgb.tif")
src1 = rasterio.plot.reshape_as_image(src.read())
#segments = quickshift(image, ratio=1, kernel_size=20, max_dist=80,
           #return_tree=False, sigma=0, convert2lab=True, random_seed=42


def onclick(event):
    print(event)
    global ix, iy
    global area
    area = 7
    ix, iy = event.xdata, event.ydata


    print ('ix ',ix)
    print ("iy ",iy)

    #X =  '${}$'.format(ix)
    #Y =  '${}$'.format(iy)
    datacursor(bbox=dict(fc='white'),formatter="longitude:{x:.2f}\nlatitude:{y:.2f}\ncolor:{z}".format)

    return
    #print ("color ",image[int(event.xdata)][int(event.ydata)])

fig,ax = plt.subplots()
graph1 = show(src.read(), transform=src.transform , ax  = ax)
fig.canvas.mpl_connect('button_press_event', onclick)
plt.show()

Answer 1

Suppose im = plt.imshow(data) is an image you plot. Then, if data is a numpy array from a three channel tif image, the RGB values of coordinates i,j is data[j,i,:] . If data is a single channel image, it would be plotted with a colormap. In that case, the RGB value is im.cmap(im.norm(data[j,i])) .

Answer 2

If you want to retrieve the RGB-value of a pixel you could use the following:

def onclick(event):
   r, g, b = event.inaxes.get_images()[0].get_cursor_data(event)[:3]
   print([r, g, b])

This way you don't need to use complicated indexing and you can also zoom-in and find the RGB-values