/Python/éliminer le background image python

éliminer le background image python

import cv2
import numpy as np

#== Parameters =======================================================================
BLUR = 21
CANNY_THRESH_1 = 10
CANNY_THRESH_2 = 200
MASK_DILATE_ITER = 10
MASK_ERODE_ITER = 10
MASK_COLOR = (0.0,0.0,1.0) # In BGR format


#== Processing =======================================================================

#-- Read image -----------------------------------------------------------------------
img = cv2.imread('C:/Temp/person.jpg')
gray = cv2.cvtColor(img,cv2.COLOR_BGR2GRAY)

#-- Edge detection -------------------------------------------------------------------
edges = cv2.Canny(gray, CANNY_THRESH_1, CANNY_THRESH_2)
edges = cv2.dilate(edges, None)
edges = cv2.erode(edges, None)

#-- Find contours in edges, sort by area ---------------------------------------------
contour_info = []
_, contours, _ = cv2.findContours(edges, cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE)
# Previously, for a previous version of cv2, this line was: 
#  contours, _ = cv2.findContours(edges, cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE)
# Thanks to notes from commenters, I've updated the code but left this note
for c in contours:
    contour_info.append((
        c,
        cv2.isContourConvex(c),
        cv2.contourArea(c),
    ))
contour_info = sorted(contour_info, key=lambda c: c[2], reverse=True)
max_contour = contour_info[0]

#-- Create empty mask, draw filled polygon on it corresponding to largest contour ----
# Mask is black, polygon is white
mask = np.zeros(edges.shape)
cv2.fillConvexPoly(mask, max_contour[0], (255))

#-- Smooth mask, then blur it --------------------------------------------------------
mask = cv2.dilate(mask, None, iterations=MASK_DILATE_ITER)
mask = cv2.erode(mask, None, iterations=MASK_ERODE_ITER)
mask = cv2.GaussianBlur(mask, (BLUR, BLUR), 0)
mask_stack = np.dstack([mask]*3)    # Create 3-channel alpha mask

#-- Blend masked img into MASK_COLOR background --------------------------------------
mask_stack  = mask_stack.astype('float32') / 255.0          # Use float matrices, 
img         = img.astype('float32') / 255.0                 #  for easy blending

masked = (mask_stack * img) + ((1-mask_stack) * MASK_COLOR) # Blend
masked = (masked * 255).astype('uint8')                     # Convert back to 8-bit 

cv2.imshow('img', masked)                                   # Display
cv2.waitKey()

#cv2.imwrite('C:/Temp/person-masked.jpg', masked)           # Save

Source: stackoverflow.com

Similar codes

3 votes247 views2 answers
Swap list items in python
Python
19 votes251 views4 answers
Python save figure
Python
3 votes281 views1 answer
Python more order of columns
Python
0 vote257 views1 answer
RuntimeError: Tensor for 'out' is on CPU, Tensor for argument #1 'self' is on CPU, but expected them to be on GPU
Python
0 vote243 views1 answer
How to save date in cookie Python
Python
0 vote262 views1 answer
Python generator in while loop
Python
0 vote264 views3 answers
Waitress
Python
4 votes263 views1 answer
Sort dict by value
Python
0 vote213 views1 answer
Dataframe pandas filter rows by list of values
Python
78 votes258 views5 answers
Join function in python
Python