overlay a smaller image on a larger image python OpenCv

Question

Hi I am creating a program that replaces a face in a image with someone else s face  However  I am stuck on trying to insert the new face into the original  larger image  I have researched ROI and addWeight needs the images to be the same size  but I haven t found a way to do this in python  Any advise is great  I am new to opencv   I am using the following test images    smaller image     larger image     Here is my Code so far    a mixer of other samples   import cv2 import cv2 cv as cv import sys import numpy  def detect img  cascade       rects   cascade detectMultiScale img  scaleFactor 1 1  minNeighbors 3  minSize  10  10   flags   cv CV HAAR SCALE IMAGE      if len rects     0          return        rects   2      rects    2      return rects  def draw rects img  rects  color       for x1  y1  x2  y2 in rects          cv2 rectangle img   x1  y1    x2  y2   color  2   if   name         main         if len sys argv     2                                             Check for error in usage syntax      print  Usage   python faces py  lt image file gt    else      img   cv2 imread sys argv 1  cv2 CV LOAD IMAGE COLOR      Read image file      if  img    None                                                print  Could not open or find the image      else          cascade   cv2 CascadeClassifier  haarcascade frontalface alt xml           gray   cv2 cvtColor img  cv CV BGR2GRAY          gray   cv2 equalizeHist gray           rects   detect gray  cascade              Extract face coordinates                  x1   rects 0  3          y1   rects 0  0          x2   rects 0  4          y2   rects 0  5          y y2-y1         x x2-x1            Extract face ROI         faceROI   gray x1 x2  y1 y2              Show face ROI         cv2 imshow  Display face ROI   faceROI          small   cv2 imread  average face png  cv2 CV LOAD IMAGE COLOR            print  here          small cv2 resize small   x  y           cv2 namedWindow  Display image               create window for display         cv2 imshow  Display image   small              Show image in the window          print  size of image     img shape           print size of image         cv2 waitKey 1000

User · Answer

Here it is   def put4ChannelImageOn4ChannelImage back  fore  x  y       rows  cols  channels   fore shape         trans indices   fore     3     0   Where not transparent     overlay copy   back y y rows  x x cols       overlay copy trans indices    fore trans indices      back y y rows  x x cols    overlay copy   test background   np zeros  1000  1000  4   np uint8  background       127  127  127  1  overlay   cv2 imread  imagee png   cv2 IMREAD UNCHANGED  put4ChannelImageOn4ChannelImage background  overlay  5  5

User · Answer

Using  fireant s idea  I wrote up a function to handle overlays  This works well for any position argument  including negative positions    def overlay image alpha img  img overlay  pos  alpha mask          Overlay img overlay on top of img at the position specified by     pos and blend using alpha mask       Alpha mask must contain values within the range  0  1  and be the     same size as img overlay               x  y   pos        Image ranges     y1  y2   max 0  y   min img shape 0   y   img overlay shape 0       x1  x2   max 0  x   min img shape 1   x   img overlay shape 1          Overlay ranges     y1o  y2o   max 0  -y   min img overlay shape 0   img shape 0  - y      x1o  x2o   max 0  -x   min img overlay shape 1   img shape 1  - x         Exit if nothing to do     if y1  gt   y2 or x1  gt   x2 or y1o  gt   y2o or x1o  gt   x2o          return      channels   img shape 2       alpha   alpha mask y1o y2o  x1o x2o      alpha inv   1 0 - alpha      for c in range channels           img y1 y2  x1 x2  c     alpha   img overlay y1o y2o  x1o x2o  c                                    alpha inv   img y1 y2  x1 x2  c     Usage is   overlay image alpha img large                      img small       0 3                        x  y                       img small       3    255 0

User · Answer

A simple 4on4 pasting function that works- def paste background foreground pos  0 0         get position and crop pasting area if needed     x   pos 0      y   pos 1      bgWidth   background shape 0      bgHeight   background shape 1      frWidth   foreground shape 0      frHeight   foreground shape 1      width   bgWidth-x     height   bgHeight-y     if frWidth lt width          width   frWidth     if frHeight lt height          height   frHeight       normalize alpha channels from 0-255 to 0-1     alpha background   background x x width y y height 3    255 0     alpha foreground   foreground  width  height 3    255 0       set adjusted colors     for color in range 0  3           fr   alpha foreground   foreground  width  height color          bg   alpha background   background x x width y y height color     1 - alpha foreground          background x x width y y height color    fr bg       set adjusted alpha and denormalize back to 0-255     background x x width y y height 3     1 -  1 - alpha foreground     1 - alpha background     255     return background

User · Answer

When attempting to write to the destination image using any of these answers above and you get the following error   ValueError  assignment destination is read-only   A quick potential fix is to set the WRITEABLE flag to true   img setflags write 1

User · Answer

A simple way to achieve what you want   import cv2 s img   cv2 imread  smaller image png   l img   cv2 imread  larger image jpg   x offset y offset 50 l img y offset y offset s img shape 0   x offset x offset s img shape 1     s img     Update  I suppose you want to take care of the alpha channel too  Here is a quick and dirty way of doing so   s img   cv2 imread  smaller image png   -1   y1  y2   y offset  y offset   s img shape 0  x1  x2   x offset  x offset   s img shape 1   alpha s   s img       3    255 0 alpha l   1 0 - alpha s  for c in range 0  3       l img y1 y2  x1 x2  c     alpha s   s img       c                                  alpha l   l img y1 y2  x1 x2  c

User · Answer

For just add an alpha channel to s img I just use cv2 addWeighted before the line l img y offset y offset s img shape 0   x offset x offset s img shape 1     s img  as following  s img cv2 addWeighted l img y offset y offset s img shape 0   x offset x offset s img shape 1   0 5 s img 0 5 0

User · Answer

Based on fireant s excellent answer above  here is the alpha blending but a bit more human legible  You may need to swap 1 0-alpha and alpha depending on which direction you re merging  mine is swapped from fireant s answer    o     s img   b     b img    for c in range 0 3       alpha   s img oy oy height  ox ox width  3    255 0     color   s img oy oy height  ox ox width  c     1 0-alpha      beta    l img by by height  bx bx width  c     alpha       l img by by height  bx bx width  c    color   beta

User · Answer

A simple function that blits an image front onto an image back and returns the result  It works with both 3 and 4-channel images and deals with the alpha channel  Overlaps are handled as well  The output image has the same size as back  but always 4 channels   The output alpha channel is given by  u v   1 uv  where u v are the alpha channels of the front and back image and -1  lt   u v  lt   1  Where there is no overlap with front  the alpha value from back is taken  import cv2  def merge image back  front  x y         convert to rgba     if back shape 2     3          back   cv2 cvtColor back  cv2 COLOR BGR2BGRA      if front shape 2     3          front   cv2 cvtColor front  cv2 COLOR BGR2BGRA         crop the overlay from both images     bh bw   back shape  2      fh fw   front shape  2      x1  x2   max x  0   min x fw  bw      y1  y2   max y  0   min y fh  bh      front cropped   front y1-y y2-y  x1-x x2-x      back cropped   back y1 y2  x1 x2       alpha front   front cropped     3 4    255     alpha back   back cropped     3 4    255            replace an area in result with overlay     result   back copy       print f af   alpha front shape  nab   alpha back shape  nfront cropped   front cropped shape  nback cropped   back cropped shape        result y1 y2  x1 x2   3    alpha front   front cropped      3    alpha back   back cropped      3      result y1 y2  x1 x2  3 4     alpha front   alpha back     1   alpha front alpha back    255      return result

[python] overlay a smaller image on a larger image python OpenCv

Examples related to python

Examples related to opencv

Examples related to object-detection