CV2. Threshold(), CV2. Findcontours(), CV2. Findcontours image contour processing

Parameters and meaning cv2.threshold

threshold(src, thresh, maxval, type[, dst]) -> retval, dst, The first return value is ** threshold ** Of float type , The second return value is the result of image pixel processing .

cv2.threshold(img,threshold,maxval,type)
img: It has to be grayscale , yes 8-bit or 32float Type ,
threshold: Judgment threshold of pixels
dst： The resulting image is the same size and channel as the original image ,
**maxval：** Maximum threshold , stay THERSH_BINARY and THERSH_BINARY_INV In the mode of

Here are the changes of pixel values in different modes .

cv2.findContours(), Find the edge outline of the binary image

cv2.findContours(image, mode, method[, contours[, hierarchy[, offset ]]]) → image, contours, hierarchy

img: The original image is a 8-bit single-channel image
mode : Represents different algorithms for obtaining contours
contours： It represents the vector of multiple contours in the image , Each contour is composed of a vector pointer
Return value ,
image - Processed image
contours - The point set of the detected contour
hierarchy - Index of each layer outline

void cv::findContours	(	InputOutputArray 	image,
OutputArrayOfArrays 	contours,
OutputArray 	hierarchy,
int 	mode,
int 	method,
Point 	offset = Point() 
)	

The retrieval methods of contour include the following ：

Reference resources ：https://www.csdn.net/tags/NtzakgxsODI4NDQtYmxvZwO0O0OO0O0O.html
Contour retrieval method
mode meaning
cv2.RETR_EXTERNAL Only detect Outline Information
cv2.RETR_LIST Between the output contours No hierarchical relationship , With list Form output outline Information
cv2.RETR_CCOMP Output Two layers of contour information , That is, the inner and outer boundaries . The upper layer is the outer boundary , The inner layer is the boundary information of the inner hole
cv2.RETR_TREE Output contour information in tree structure
Contour approximation
method meaning
cv2.CHAIN_APPROX_NONE Store all boundary points
cv2.CHAIN_APPROX_SIMPLE Compress vertical 、 level 、 Diagonally , Keep only endpoints
cv2.CHAIN_APPROX_TX89_L1 Use teh-Chini The approximate algorithm
cv2.CHAIN_APPROX_TC89_KCOS Use teh-Chini The approximate algorithm

cv2.findContours

Reference resources ：https://blog.csdn.net/Easen_Yu/article/details/89380578

cv2.drawContours(image, contours, contourIdx, color, thickness=None, lineType=None, hierarchy=None, maxLevel=None, offset=None)

**image:** Is to indicate on which image the outline is drawn ;image Three channels to show the outline
**contours:** It's the outline itself , stay Python There is a list;
contourIdx: Specifies the sketch profile list Which outline in the , If it is -1, Then draw all the contours in it . The following parameters are simple .
thickness Indicates the width of the contour line , If it is -1（cv2.FILLED）, Fill mode

Comprehensive use of examples

import cv2

img = cv2.imread('test4.jpg')
image_gray  = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
## t Is the set threshold ,binary Is the image obtained by threshold comparison 
t, binary = cv2.threshold(image_gray, 127, 255, cv2.THRESH_BINARY)

#  There are three return values , Be sure to define three return values , Otherwise, it will be wrong ,ValueError: too many values to unpack 
_,contours, hierarchy = cv2.findContours(binary, cv2.RETR_LIST, cv2.CHAIN_APPROX_NONE)

# Draw outline 
r = cv2.drawContours(img, contours, -1, (0, 255, 0), 2)
cv2.imshow('img1',r)
# cv2.imshow('res', img)
k = cv2.waitKey(0)
if k == ord('e'):
    exit(0)