Contours are curves that join all the continuous points along a boundary that have the same color or intensity. They are widely used in shape analysis, object detection, and recognition tasks in computer vision.

OpenCV provides the findContours() method to extract contours and drawContours() to display them.

Steps:

1. Convert image to grayscale
2. Apply threshold or Canny edge detection
3. Use findContours()
4. Draw the contours with drawContours()

cv::Mat img = cv::imread("shapes.png");
cv::Mat gray, edges;
cv::cvtColor(img, gray, cv::COLOR_BGR2GRAY);
cv::Canny(gray, edges, 100, 200);

std::vector<std::vector<cv::Point>> contours;
cv::findContours(edges, contours, cv::RETR_EXTERNAL, cv::CHAIN_APPROX_SIMPLE);
cv::drawContours(img, contours, -1, cv::Scalar(0, 255, 0), 2);

cv::findContours() signature:

void findContours(InputOutputArray image, 
                  OutputArrayOfArrays contours,
                  OutputArray hierarchy,
                  int mode,
                  int method);

Notes:

• mode - Retrieval method (e.g., RETR_EXTERNAL, RETR_TREE)
• method - Approximation (e.g., CHAIN_APPROX_SIMPLE, CHAIN_APPROX_NONE)

Use drawContours() to visualize contours. You can specify:

• All contours: -1 as index
• Thickness of lines
• Color in BGR format

cv::drawContours(image, contours, -1, cv::Scalar(255, 0, 0), 2);

Hierarchy describes parent-child relationships between contours. It is returned as a vector with the format:

[Next, Previous, First_Child, Parent]

std::vector<cv::Vec4i> hierarchy;
cv::findContours(edges, contours, hierarchy, cv::RETR_TREE, cv::CHAIN_APPROX_SIMPLE);

You can calculate properties using:

• cv::arcLength() – perimeter
• cv::contourArea() – area
• cv::approxPolyDP() – polygonal approximation
• cv::boundingRect() – rectangular bound
• cv::minEnclosingCircle() – enclosing circle

double peri = cv::arcLength(contours[i], true);
std::vector<cv::Point> approx;
cv::approxPolyDP(contours[i], approx, 0.02 * peri, true);
cv::Rect rect = cv::boundingRect(approx);

Contour Filtering: Filter by size, shape, or hierarchy:

for (size_t i = 0; i < contours.size(); i++) {
  double area = cv::contourArea(contours[i]);
  if (area > 1000) {
    cv::drawContours(image, contours, (int)i, cv::Scalar(0, 255, 0), 2);
  }
}

Shape Matching:

double match = cv::matchShapes(contour1, contour2, cv::CONTOURS_MATCH_I1, 0);

Contours provide a flexible foundation for many image processing tasks. Mastering their extraction and manipulation with OpenCV in C++ opens up advanced image analysis possibilities such as object recognition, measurement, tracking, and shape matching.