Extracting foreground image as mask by thresholding - python

I'm looking for a robust way to extract the foreground from an image where the background has some noise in it.
So, the image I want to use it on is:
My attempt was to use the Otsu thresholding. I did that in Python as follows:
from skimage.filter import threshold_otsu
import os.path as path
import matplotlib.pyplot as plt
img = io.imread(path.expanduser('~/Desktop/62.jpg'))
r_t = threshold_otsu(img[:, :, 0])
g_t = threshold_otsu(img[:, :, 1])
b_t = threshold_otsu(img[:, :, 2])
m = np.zeros((img.shape[0], img.shape[1]), dtype=np.uint8)
mask = (img[:, :, 0] < r_t) & (img[:, :, 1] < g_t) & (img[:, :, 2] < b_t)
m[~mask] = 255
plt.imshow(m)
plt.show()
This gives the R, G, B threshold as (62 67 64), which is a bit high. The result is:
This image is also one of the images where Otsu thresholding worked best. If I use a manual threshold like a value of 30, it works quite well. The result is:
I was wondering if there are some other approaches that I should try. Segmentation really is not my area of expertise and what I can do out of the box seem limited.

Your image looks not very colorful. So you can perform the segmentation on the gray values and not on each color separately and then combining three masks.
Looking at the package scikit-image.filter there are several other threshold methods. I tried them all and found threshold_isodata to perform extremely well giving almost the same image as your desired image. Therefore I recommend the isodata algorithm.
Example:
import numpy as np
import skimage.io as io
import skimage.filter as filter
import matplotlib.pyplot as plt
img = io.imread('62.jpg')
gray = np.sum(img, axis=2) # summed up over red, green, blue
#threshold = filter.threshold_otsu(gray) # delivers very high threshold
threshold = filter.threshold_isodata(gray) # works extremely well
#threshold = filter.threshold_yen(gray) # delivers even higher threshold
print(threshold)
plt.imshow(gray > threshold)
plt.show()
gives:

Related

How to delete entries below threshold in class 'imantics.annotation.Polygons'?

I have greyscale images with features of interest displayed as grey and white, and background as black.
I am trying to draw polygons around the features of interest.
My problem is that polygons are drawn e.g. around the edge of the images as well (input image). In the code below I have tried to filter out these "false positive" features of interest using gaussian blur and morphological operations (see code below),
import cv2
import matplotlib.pyplot as plt
import numpy as np
from imantics import Polygons, Mask
import imantics as imcs
import skimage
from shapely.geometry import Polygon as Pollygon
import matplotlib.image as mpimg
import PIL
mask = cv2.imread('mask.jpg',64)
print(mask.max())
print(mask.min())
# Apply gaussian blur filter
mask = cv2.GaussianBlur(mask,(9,9),0)
mask = cv2.GaussianBlur(mask,(9,9),0)
mask = cv2.GaussianBlur(mask,(9,9),0)
mask = cv2.GaussianBlur(mask,(9,9),0)
mask = cv2.GaussianBlur(mask,(9,9),0)
ellipseFootprint = skimage.morphology.footprints.ellipse(1, 1)
squareFootprint = skimage.morphology.footprints.square(8)
maskMorph = mask
for i in range(10):
maskMorph = skimage.morphology.erosion(maskMorph, footprint=ellipseFootprint, out=None)
print(i)
for k in range(2):
maskMorph = skimage.morphology.dilation(maskMorph, footprint=None, out=None)
print(k)
polygons = Mask(maskMorph).polygons()
print(len(polygons.segmentation))
print(type(polygons))
print(polygons.segmentation)
newPoly = polygons.draw(mask, color=[255, 255, 0],
thickness=3)
cv2.imshow("title", newPoly)
cv2.waitKey()
Indeed, I have tried to "filter" out smaller features/polygons and "false positive" features of interest in images using gaussian blur filter and morphological operations, but I am struggling with getting rid of all (see output image).
My thinking is therefore to add a minimum (size) threshold for the features/polygons in the image to be kept.
I have started on the following, but am not sure how to progress.
lengthPolySeg = len(polygons.segmentation)
for l in range(lengthPolySeg-1):
if len(polygons.segmentation[l]) < 50:
Any advise would be most appreciated.

Get mask of image without using OpenCV

I'm trying the following to get the mask out of this image, but unfortunately I fail.
import numpy as np
import skimage.color
import skimage.filters
import skimage.io
# get filename, sigma, and threshold value from command line
filename = 'pathToImage'
# read and display the original image
image = skimage.io.imread(fname=filename)
skimage.io.imshow(image)
# blur and grayscale before thresholding
blur = skimage.color.rgb2gray(image)
blur = skimage.filters.gaussian(blur, sigma=2)
# perform inverse binary thresholding
mask = blur < 0.8
# use the mask to select the "interesting" part of the image
sel = np.ones_like(image)
sel[mask] = image[mask]
# display the result
skimage.io.imshow(sel)
How can I obtain the mask?
Is there a general approach that would work for this image as well. without custom fine-tuning and changing parameters?
Apply high contrast (maximum possible value)
convert to black & white image using high threshold (I've used 250)
min filter (value=8)
max filter (value=8)
Here is how you can get a rough mask using only the skimage library methods:
import numpy as np
from skimage.io import imread, imsave
from skimage.feature import canny
from skimage.color import rgb2gray
from skimage.filters import gaussian
from skimage.morphology import dilation, erosion, selem
from skimage.measure import find_contours
from skimage.draw import polygon
def get_mask(img):
kernel = selem.rectangle(7, 6)
dilate = dilation(canny(rgb2gray(img), 0), kernel)
dilate = dilation(dilate, kernel)
dilate = dilation(dilate, kernel)
erode = erosion(dilate, kernel)
mask = np.zeros_like(erode)
rr, cc = polygon(*find_contours(erode)[0].T)
mask[rr, cc] = 1
return gaussian(mask, 7) > 0.74
def save_img_masked(file):
img = imread(file)[..., :3]
mask = get_mask(img)
result = np.zeros_like(img)
result[mask] = img[mask]
imsave("masked_" + file, result)
save_img_masked('belt.png')
save_img_masked('bottle.jpg')
Resulting masked_belt.png:
Resulting masked_bottle.jpg:
One approach uses the fact that the background changes color only very slowly. Here I apply the gradient magnitude to each of the channels and compute the norm of the result, giving me an image highlighting the quicker changes in color. The watershed of this (with sufficient tolerance) should have one or more regions covering the background and touching the image edge. After identifying those regions, and doing a bit of cleanup we get these results (red line is the edge of the mask, overlaid on the input image):
I did have to adjust the tolerance, with a lower tolerance in the first case, more of the shadow is seen as object. I think it should be possible to find a way to set the tolerance based on the statistics of the gradient image, I have not tried.
There are no other parameters to tweak here, the minimum object area, 300, is quite safe; an alternative would be to keep only the one largest object.
This is the code, using DIPlib (disclaimer: I'm an author). out is the mask image, not the outline as displayed above.
import diplib as dip
import numpy as np
# Case 1:
img = dip.ImageRead('Pa9DO.png')
img = img[362:915, 45:877] # cut out actual image
img = img(slice(0,2)) # remove alpha channel
tol = 7
# Case 2:
#img = dip.ImageRead('jTnVr.jpg')
#tol = 1
# Compute gradient
gm = dip.Norm(dip.GradientMagnitude(img))
# Compute watershed with tolerance
lab = dip.Watershed(gm, connectivity=1, maxDepth=tol, flags={'correct','labels'})
# Identify regions touching the image edge
ll = np.unique(np.concatenate((
np.unique(lab[:,0]),
np.unique(lab[:,-1]),
np.unique(lab[0,:]),
np.unique(lab[-1,:]))))
# Remove regions touching the image edge
out = dip.Image(lab.Sizes(), dt='BIN')
out.Fill(1)
for l in ll:
if l != 0: # label zero is for the watershed lines
out = out - (lab == l)
# Remove watershed lines
out = dip.Opening(out, dip.SE(3, 'rectangular'))
# Remove small regions
out = dip.AreaOpening(out, filterSize=300)
# Display
dip.Overlay(img, dip.Dilation(out, 3) - out).Show()

Denoising a photo with Python

I have the following image which is a scanned copy of an old book. I want to remove the noise in the background (which is a bit reddish) that is coming due to the scanning of the old photo.
Update:
After applying opencv, following the parameter settings in opencv doc, I am getting the following output.
Please help fixing this.
The code that I am using:
import numpy as np
import cv2
from matplotlib import pyplot as plt
def display_image_in_actual_size(im_data):
dpi = 80
height, width, depth = im_data.shape
# What size does the figure need to be in inches to fit the image?
figsize = width / float(dpi), height / float(dpi)
# Create a figure of the right size with one axes that takes up the full figure
fig = plt.figure(figsize=figsize)
ax = fig.add_axes([0, 0, 1, 1])
# Hide spines, ticks, etc.
ax.axis('off')
# Display the image.
ax.imshow(im_data, cmap='gray')
plt.show()
img = cv2.imread('scan03.jpg')
dst = cv2.fastNlMeansDenoisingColored(img,None,10,10,7,21)
display_image_in_actual_size(img)
display_image_in_actual_size(dst)
The color of some pixels which has near threshold pixel values will be affected, but that depends on the task, here is one solution that you might adjust the threshold to a value that suits your task, also you might remove the median filter, or reduce the sigma value(5) if it affects the text badly, you might have some undesired noise, but the text will be readable.
import numpy as np
import matplotlib.pyplot as plt
import cv2
# Read Image
img = cv2.imread('input.jpg')
# BGR --> RGB
RGB = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
# BGR --> Gray
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
# Set thresholds
th_white = 210
th_black = 85
# copy original gray
mask_white = gray.copy()
mask_black = gray.copy()
# Thresholding
mask_white[mask_white<th_white] = 0
mask_black[mask_black<th_black] = 0
mask_white[mask_white>=th_white] = 255
mask_black[mask_black>=th_black] = 255
# Median Filtering (you can remove if the text is not readable)
median_white = cv2.medianBlur(mask_white,5)
median_black = cv2.medianBlur(mask_black,5)
# Mask 3 channels
mask_white_3 = np.stack([median_white, median_white, median_white], axis=2)
mask_black_3 = np.stack([median_black, median_black, median_black], axis=2)
# Masking the image(in RGB)
result1 = np.maximum(mask_white_3, RGB)
result2 = np.minimum(mask_black_3, result1)
# Visualize the results
plt.imshow(result2)
plt.axis('off')
plt.show()
opencv library has couple of denoisong functions.
You can find reading with examples here

Is there a better way to separate the writing from the background?

I am working on a project where I should apply and OCR on some documents.
The first step is to threshold the image and let only the writing (whiten the background).
Example of an input image: (For the GDPR and privacy reasons, this image is from the Internet)
Here is my code:
import cv2
import numpy as np
image = cv2.imread('b.jpg')
image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
h = image.shape[0]
w = image.shape[1]
for y in range(0, h):
for x in range(0, w):
if image[y, x] >= 120:
image[y, x] = 255
else:
image[y, x] = 0
cv2.imwrite('output.jpg', image)
Here is the result that I got:
When I applied pytesseract to the output image, the results were not satisfying (I know that an OCR is not perfect). Although I tried to adjust the threshold value (in this code it is equal to 120), the result was not as clear as I wanted.
Is there a way to make a better threshold in order to only keep the writing in black and whiten the rest?
After digging deep in StackOverflow questions, I found this answer which is about removing watermark using opencv.
I adapted the code to my needs and this is what I got:
import numpy as np
import cv2
image = cv2.imread('a.png')
img = image.copy()
alpha =2.75
beta = -160.0
denoised = alpha * img + beta
denoised = np.clip(denoised, 0, 255).astype(np.uint8)
#denoised = cv2.fastNlMeansDenoising(denoised, None, 31, 7, 21)
img = cv2.cvtColor(denoised, cv2.COLOR_BGR2GRAY)
h = img.shape[0]
w = img.shape[1]
for y in range(0, h):
for x in range(0, w):
if img[y, x] >= 220:
img[y, x] = 255
else:
img[y, x] = 0
cv2.imwrite('outpu.jpg', img)
Here is the output image:
The good thing about this code is that it gives good results not only with this image, but also with all the images that I tested.
I hope it helps anyone who had the same problem.
You can use adaptive thresholding. From documentation :
In this, the algorithm calculate the threshold for a small regions of the image. So we get different thresholds for different regions of the same image and it gives us better results for images with varying illumination.
import numpy as np
import cv2
image = cv2.imread('b.jpg')
image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
image = cv2.medianBlur(image ,5)
th1 = cv2.adaptiveThreshold(image,255,cv2.ADAPTIVE_THRESH_MEAN_C,\
cv2.THRESH_BINARY,11,2)
th2 = cv2.adaptiveThreshold(image,255,cv2.ADAPTIVE_THRESH_GAUSSIAN_C,\
cv2.THRESH_BINARY,11,2)
cv2.imwrite('output1.jpg', th1 )
cv2.imwrite('output2.jpg', th2 )

Filling holes in image with OpenCV or Skimage

I m trying to fill holes for a chessboard for stereo application. The chessboard is at micro scale thus it is complicated to avoid dust... as you can see :
Thus, the corners detection is impossible. I tried with SciPy's binary_fill_holes or similar approaches but i have a full black image, i dont understand.
Here is a function that replaces the color of each pixel with the color that majority of its neighbor pixels have.
import numpy as np
import cv2
def remove_noise(gray, num):
Y, X = gray.shape
nearest_neigbours = [[
np.argmax(
np.bincount(
gray[max(i - num, 0):min(i + num, Y), max(j - num, 0):min(j + num, X)].ravel()))
for j in range(X)] for i in range(Y)]
result = np.array(nearest_neigbours, dtype=np.uint8)
cv2.imwrite('result2.jpg', result)
return result
Demo:
img = cv2.imread('mCOFl.png')
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
remove_noise(gray, 10)
Input image:
Out put:
Note: Since this function replace the color of corner pixels too, you can sue cv2.goodFeaturesToTrack function to find the corners and restrict the denoising for that pixels
corners = cv2.goodFeaturesToTrack(gray, 100, 0.01, 30)
corners = np.squeeze(np.int0(corners))
You can use morphology: dilate, and then erode with same kernel size.
A faster, more accurate way is to use skimage.morphology.remove_small_objects docs
im = imread('a.png',cv2.IMREAD_GRAYSCALE)
im = im ==255
from skimage import morphology
cleaned = morphology.remove_small_objects(im, 200)

Categories