Opencv learning -- arithmetic operation of image of basic operation

The addition of images

Adding two images together can use OpenCV Of cv2.add() function , Or simply pass numpy Add two images , Such as res=img1+img2.

The added two images should have Same size and type , Or the second image can be a scalar value .

OpenCV Add and Numpy Between addition Differences exist .OpenCV The addition of is a saturation operation ,Numpy The addition of is a modular operation .

Explanation of differences ：

x=np.uint2([250])
y=np.uint8([10])

print(cv2.add(x, y))        #250+10=260=>255    OpenCV Add      Saturation operation 

print(x+y)                #250+10=260%256=4    Numpy Add      Modular arithmetic 

When adding two images , OpenCV The result will be better , therefore Use as much as possible OpenCV The function in .

The blending of images

Image blending is actually addition , But the difference is between the two images Different weights , It gives a sense of mixing or transparency . The calculation formula of image blending is ：

g(x)=(1-a)*f0(x)+a*f1(x) 

By modifying the a Value （0-1）, Different mixing effects can be achieved .

function cv2.addWeighted() You can mix pictures according to the following formula ：

dst=a*img1+b*img2+c    # here c take 0

Image blending function ：cv2.addWeighted()

The function prototype ： cv2.addWeighted(src1,alpha,src2,beta,c)

src1,src2： Two images that need to be mixed

alpha,beta： The weight , Add to 1

c： The amount of brightness adjustment , Be sure to write , Can write 0, That is, the brightness is not adjusted .