Python OpenCV: Crop image to contents, and make background transparent - python

I have the following image:
I want to crop the image to the actual contents, and then make the background (the white space behind) transparent. I have seen the following question: How to crop image based on contents (Python & OpenCV)?, and after looking at the answer, and trying it, I got the following code:
img = cv.imread("tmp/"+img+".png")
mask = np.zeros(img.shape[:2],np.uint8)
bgdModel = np.zeros((1,65),np.float64)
fgdModel = np.zeros((1,65),np.float64)
rect = (55,55,110,110)
cv.grabCut(img,mask,rect,bgdModel,fgdModel,5,cv.GC_INIT_WITH_RECT)
mask2 = np.where((mask==2)|(mask==0),0,1).astype('uint8')
img = img*mask2[:,:,np.newaxis]
plt.imshow(img),plt.colorbar(),plt.show()
But when I try this code, I get the following result:
Which isn't really the result I'm searching for, expected result:

Here is one way to do that in Python/OpenCV.
As I mentioned in my comment, your provided image has a white circle around the cow and then a transparent background. I have made the background fully white as my input.
Input:
import cv2
import numpy as np
# read image
img = cv2.imread('cow.png')
# convert to grayscale
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
# invert gray image
gray = 255 - gray
# threshold
thresh = cv2.threshold(gray,0,255,cv2.THRESH_BINARY)[1]
# apply close and open morphology to fill tiny black and white holes and save as mask
kernel = np.ones((3,3), np.uint8)
mask = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)
mask = cv2.morphologyEx(mask, cv2.MORPH_OPEN, kernel)
# get contours (presumably just one around the nonzero pixels)
contours = cv2.findContours(mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
contours = contours[0] if len(contours) == 2 else contours[1]
cntr = contours[0]
x,y,w,h = cv2.boundingRect(cntr)
# make background transparent by placing the mask into the alpha channel
new_img = cv2.cvtColor(img, cv2.COLOR_BGR2BGRA)
new_img[:, :, 3] = mask
# then crop it to bounding rectangle
crop = new_img[y:y+h, x:x+w]
# save cropped image
cv2.imwrite('cow_thresh.png',thresh)
cv2.imwrite('cow_mask.png',mask)
cv2.imwrite('cow_transparent_cropped.png',crop)
# show the images
cv2.imshow("THRESH", thresh)
cv2.imshow("MASK", mask)
cv2.imshow("CROP", crop)
cv2.waitKey(0)
cv2.destroyAllWindows()
Thresholded image:
Mask:
Cropped result with transparent background:

Given that the background to be converted to transparent has its BGR channels white (like in your image), you can do:
import cv2
import numpy as np
img = cv2.imread("cat.png")
img[np.where(np.all(img == 255, -1))] = 0
img_transparent = cv2.cvtColor(img, cv2.COLOR_BGR2BGRA)
img_transparent[np.where(np.all(img == 0, -1))] = 0
cv2.imshow("transparent.png", img_transparent)
Input image:
Output image:
We can tell the second image is transparent by clicking on it; the transparent background will show up a grey (in Firefox, at least).

What worked to me is:
original_image = cv2.imread(path)
#Converting the bgr image to an image with the alpha channel
original_image = cv2.cvtColor(original_image, cv2.BGR2BGRA)
#Transforming every alpha pixel to a transparent pixel.
original_image[np.where(np.all(original_image == 255, -1))] = 0
And then writing the image.

Related

Identify and count objects different from background

I try to use python, NumPy, and OpenCV to analyze the image below and just draw a circle on each object found. The idea here is not to identify the bug only identify any object that is different from the background.
Original Image:
Here is the code that I'm using.
import cv2
import numpy as np
img = cv2.imread('per.jpeg', cv2.IMREAD_GRAYSCALE)
if cv2.__version__.startswith('2.'):
detector = cv2.SimpleBlobDetector()
else:
detector = cv2.SimpleBlobDetector_create()
keypoints = detector.detect(img)
print(len(keypoints))
imgKeyPoints = cv2.drawKeypoints(img, keypoints, np.array([]), (0,0,255), cv2.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS)
status = cv2.imwrite('teste.jpeg',imgKeyPoints)
print("Image written to file-system : ",status)
But the problem is that I'm getting only a greyscale image as result without any counting or red circle, as shown below:
Since I'm new to OpenCV and object recognition world I'm not able to identify what is wrong, and any help will be very appreciated.
Here is one way in Python/OpenCV.
Threshold on the bugs color in HSV colorspace. Then use morphology to clean up the threshold. Then get contours. Then find the minimum enclosing circle around each contour. Then bias the radius to make a bit larger and draw the circle around each bug.
Input:
import cv2
import numpy as np
# read image
img = cv2.imread('bugs.jpg')
# convert image to hsv colorspace
hsv = cv2.cvtColor(img, cv2.COLOR_BGR2HSV)
# threshold on bugs color
lower=(0,90,10)
upper=(100,250,170)
thresh = cv2.inRange(hsv, lower, upper)
# apply morphology to clean up
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (3,3))
morph = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (6,6))
morph = cv2.morphologyEx(morph, cv2.MORPH_CLOSE, kernel)
# get external contours
contours = cv2.findContours(morph, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
contours = contours[0] if len(contours) == 2 else contours[1]
result = img.copy()
bias = 10
for cntr in contours:
center, radius = cv2.minEnclosingCircle(cntr)
cx = int(round(center[0]))
cy = int(round(center[1]))
rr = int(round(radius)) + bias
cv2.circle(result, (cx,cy), rr, (0, 0, 255), 2)
# save results
cv2.imwrite('bugs_threshold.jpg', thresh)
cv2.imwrite('bugs_cleaned.jpg', morph)
cv2.imwrite('bugs_circled.jpg', result)
# display results
cv2.imshow('thresh', thresh)
cv2.imshow('morph', morph)
cv2.imshow('result', result)
cv2.waitKey(0)
cv2.destroyAllWindows()
Threshold Image:
Morphology Cleaned Image:
Resulting Circles:

Removing Border from the Image

I have a set of images like below, where each image consist of the actual image part with the black borders of varying sizes around it. My objective is to remove those borders and just extract the image, which later can be used for other tasks.
I have tried using canny edge detector, to identify the border and get only the image out of it, however its not working as expected.
I have also tried techniques mentioned in other posts like below, but these too doesnot work in my case.
In [1]: from PIL import Image, ImageChops
In [3]: im = Image.open('iI3ZE.jpg')
In [4]: def trim(im):
...: bg = Image.new(im.mode, im.size, im.getpixel((0,0)))
...: diff = ImageChops.difference(im, bg)
...: diff = ImageChops.add(diff, diff, 2.0, -100)
...: bbox = diff.getbbox()
...: if bbox:
...: return im.crop(bbox)
Any leads for the above query would be helpful.
Here is one way in Python/OpenCV. Read the image as grayscale. Then threshold at zero since background is black. Then apply morphology open to clean it up. Then find contours and get the bounding box of the largest one. Then crop using the bounding box.
Input:
import cv2
import numpy as np
# load image
img = cv2.imread("sonar.png", cv2.IMREAD_GRAYSCALE)
# threshold
thresh = cv2.threshold(img, 0, 255, cv2.THRESH_BINARY)[1]
# apply open morphology
#kernel = np.ones((5,5), np.uint8)
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (15,15))
morph = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
# get bounding box coordinates from largest external contour
contours = cv2.findContours(morph, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
contours = contours[0] if len(contours) == 2 else contours[1]
big_contour = max(contours, key=cv2.contourArea)
x,y,w,h = cv2.boundingRect(big_contour)
# crop image contour image
result = img.copy()
result = result[y:y+h, x:x+w]
# write result to disk
cv2.imwrite("sonar_thresh.jpg", thresh)
cv2.imwrite("sonar_morph.jpg", morph)
cv2.imwrite("sonar_cropped.png", result)
# display results
cv2.imshow("THRESH", thresh)
cv2.imshow("MORPH", morph)
cv2.imshow("CROPPED", result)
cv2.waitKey(0)
cv2.destroyAllWindows()
Threshold image:
Morphology cleaned image:
Cropped image:

reduce shape of image by removing whiteness arround

from pdf2image import convert_from_path
import cv2,numpy,os
def pil_to_cv2(image):
open_cv_image = numpy.array(image)
return open_cv_image[:, :, ::-1].copy()
images = convert_from_path('test.pdf')
cv_h=[pil_to_cv2(i) for i in images]
for img in cv_h:
#function_to_crop()
cv2.imwrite('modified.png', img)
How can I remove the extra whiteness from the image (top,sideways,under) without actually intercepting the drawing, The drawings from pdf are from different sizes so I can't crop the images by a fixed number.
Ideally,the output would look like this
Here is another way to do that in Python/OpenCV.
Read the image
Convert to gray and invert the polarity
Threshold
Apply morphology close to fill in holes and make one solid region
Get the outer contour and its bounding box
Use the bounding box to crop the image using Numpy slicing
Save the result
Input:
import cv2
import numpy as np
# read image
img = cv2.imread('multipower.png')
# convert to grayscale
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
# invert gray image
gray = 255 - gray
# threshold
thresh = cv2.threshold(gray,0,255,cv2.THRESH_BINARY)[1]
# apply close and open morphology
kernel = np.ones((75,75), np.uint8)
mask = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)
# get contours (presumably just one around the nonzero pixels)
contours = cv2.findContours(mask, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
contours = contours[0] if len(contours) == 2 else contours[1]
cntr = contours[0]
x,y,w,h = cv2.boundingRect(cntr)
# draw contour on input
contour_img = img.copy()
cv2.drawContours(contour_img,[cntr],0,(0,0,255),3)
# crop to bounding rectangle
crop = img[y:y+h, x:x+w]
# save cropped image
cv2.imwrite('multipower_thresh.png',thresh)
cv2.imwrite('multipower_mask.png',mask)
cv2.imwrite('multipower_contour.png',contour_img)
cv2.imwrite('multipower_cropped.png',crop)
# show the images
cv2.imshow("THRESH", thresh)
cv2.imshow("MASK", mask)
cv2.imshow("CONTOUR", contour_img)
cv2.imshow("CROP", crop)
cv2.waitKey(0)
cv2.destroyAllWindows()
Thresholded Image:
Morphology closed image:
Contour image:
Result:
import cv2 as cv
import numpy as np
frame = cv.imread('7dcoI.png')
frame_gray = cv.cvtColor(frame, cv.COLOR_BGR2GRAY)
mask=cv.threshold(frame_gray, 85, 255, cv.THRESH_BINARY )[1]
rows, cols = mask.shape
non_empty_columns = np.where(mask.min(axis=0)==0)[0]
non_empty_rows = np.where(mask.min(axis=1)==0)[0]
cropBox = (min(non_empty_rows), min(max(non_empty_rows), rows), min(non_empty_columns), min(max(non_empty_columns), cols))
cropped = frame[cropBox[0]:cropBox[1]+1, cropBox[2]:cropBox[3]+1 , :]
cv.imwrite('out_mask.png', cropped)

Finding different boundaries around a set of points in an image

I have an image which looks something like this:
here is the image
I need to make borders around the points such that they are divided into clusters. For example, the center of the image is one region. One other region can be the top of the image. How can I achieve this, with python preferably?
Here is one way to do that in Python/OpenCV.
- Read the input as unchanged, since it has transparency
- Separate the base image and the alpha channel
- Mask the base image with the alpha channel so as to make the white outer region with the text into all black
- Convert that image into grayscale and then into black/white
- Apply morphology close to connect all the dots in the regions
- Find all contours larger than some minimum area
- Draw the contours on the base image
- Save the results
Input:
import cv2
import numpy as np
# read image with transparency
image = cv2.imread("dots.png", cv2.IMREAD_UNCHANGED)
# separate base image and alpha channel and make background under transparency into black to remove white border and text
base = image[:,:,0:3]
alpha = image[:,:,3]
alpha = cv2.merge([alpha,alpha,alpha])
img = cv2.bitwise_and(base, alpha)
# convert img to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
#do threshold on gray image
thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY)[1]
# apply morphology close
kernel = cv2.getStructuringElement(cv2.MORPH_RECT, (15, 15))
close = cv2.morphologyEx(thresh, cv2.MORPH_CLOSE, kernel)
# Get contours
cnts = cv2.findContours(close, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
cnts = cnts[0] if len(cnts) == 2 else cnts[1]
result = base.copy()
for c in cnts:
area = cv2.contourArea(c)
if area > 100:
cv2.drawContours(result, [c], -1, (0, 255, 0), 1)
# display it
cv2.imshow("BASE", base)
cv2.imshow("BLACKENED", img)
cv2.imshow("CLOSED", close)
cv2.imshow("RESULT", result)
cv2.waitKey(0)
# write results to disk
cv2.imwrite("dots_blackened.png", img)
cv2.imwrite("dots_closed.png", close)
cv2.imwrite("dots_clusters.png", result)
Base Image with transparency blackened:
Morphology Close Image:
Contours on base image:

find rectangle in image and extract text inside of it to save it as new image

I am new to OpenCV so I really need your help. I have a bunch of images like this one:
I need to detect the rectangle on the image, extract the text part from it and save it as a new image.
Can you please help me with this?
Thank you!
Just to add to Danyals answer I have added an example code with steps written in comments. For this image you don't even need to perform morphological opening on the image. But usually for this kind of noise in the image it is recomended. Cheers!
import cv2
import numpy as np
# Read the image and create a blank mask
img = cv2.imread('napis.jpg')
h,w = img.shape[:2]
mask = np.zeros((h,w), np.uint8)
# Transform to gray colorspace and invert Otsu threshold the image
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)
_, thresh = cv2.threshold(gray,0,255,cv2.THRESH_BINARY_INV+cv2.THRESH_OTSU)
# ***OPTIONAL FOR THIS IMAGE
### Perform opening (erosion followed by dilation)
#kernel = np.ones((2,2),np.uint8)
#opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
# ***
# Search for contours, select the biggest and draw it on the mask
_, contours, hierarchy = cv2.findContours(thresh, # if you use opening then change "thresh" to "opening"
cv2.RETR_TREE,cv2.CHAIN_APPROX_NONE)
cnt = max(contours, key=cv2.contourArea)
cv2.drawContours(mask, [cnt], 0, 255, -1)
# Perform a bitwise operation
res = cv2.bitwise_and(img, img, mask=mask)
########### The result is a ROI with some noise
########### Clearing the noise
# Create a new mask
mask = np.zeros((h,w), np.uint8)
# Transform the resulting image to gray colorspace and Otsu threshold the image
gray = cv2.cvtColor(res,cv2.COLOR_BGR2GRAY)
_, thresh = cv2.threshold(gray,0,255,cv2.THRESH_BINARY+cv2.THRESH_OTSU)
# Search for contours and select the biggest one again
_, contours, hierarchy = cv2.findContours(thresh,cv2.RETR_TREE,cv2.CHAIN_APPROX_NONE)
cnt = max(contours, key=cv2.contourArea)
# Draw it on the new mask and perform a bitwise operation again
cv2.drawContours(mask, [cnt], 0, 255, -1)
res = cv2.bitwise_and(img, img, mask=mask)
# If you will use pytesseract it is wise to make an aditional white border
# so that the letters arent on the borders
x,y,w,h = cv2.boundingRect(cnt)
cv2.rectangle(res,(x,y),(x+w,y+h),(255,255,255),1)
# Crop the result
final_image = res[y:y+h+1, x:x+w+1]
# Display the result
cv2.imshow('img', final_image)
cv2.waitKey(0)
cv2.destroyAllWindows()
Result:
One way to do this (if the rectangle sizes are somewhat predictable) is:
Convert the image to black and white
Invert the image
Perform morphological opening on the image from (2) with a horizontal line / rectangle (I tried with 2x30).
Perform morphological opening on the image from (2) with a vertical line (I tried it with 15x2).
Add the images from (3) and (4). You should only have a white rectangle now. Now can remove all corresponding rows and columns in the original image that are entirely zero in this image.

Categories

Resources