How to Use Contour Detection in OpenCV for Computer Vision

Use cv2.findContours() in OpenCV to detect contours by passing a binary image. Contours represent boundaries of objects and can be processed for shape analysis or object detection.

Syntax

The main function for contour detection in OpenCV is cv2.findContours(). It takes a binary image and returns contours and hierarchy information.

• image: The input binary image (usually after thresholding or edge detection).
• mode: Contour retrieval mode (e.g., cv2.RETR_EXTERNAL for outer contours).
• method: Contour approximation method (e.g., cv2.CHAIN_APPROX_SIMPLE to compress horizontal, vertical, and diagonal segments).

python

contours, hierarchy = cv2.findContours(image, mode, method)

Example

This example loads an image, converts it to grayscale, applies thresholding to get a binary image, detects contours, and draws them on the original image.

python

import cv2

# Load image
image = cv2.imread('shapes.png')

# Convert to grayscale
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)

# Apply binary threshold
_, thresh = cv2.threshold(gray, 127, 255, cv2.THRESH_BINARY)

# Find contours
contours, hierarchy = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

# Draw contours on the original image
cv2.drawContours(image, contours, -1, (0, 255, 0), 3)

# Save the result
cv2.imwrite('contours_output.png', image)

# Print number of contours found
print(f'Number of contours found: {len(contours)}')

Output

Number of contours found: 3

Common Pitfalls

• Passing a non-binary image to cv2.findContours() will not detect contours correctly. Always threshold or edge-detect first.
• Using the wrong retrieval mode can miss inner contours or hierarchy details.
• Not handling the return values correctly: in OpenCV 4.x, findContours returns contours, hierarchy, but in older versions it returns image, contours, hierarchy.

python

import cv2

# Wrong: passing color image directly
image = cv2.imread('shapes.png')
contours, hierarchy = cv2.findContours(image, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)  # This will fail

# Right: convert to grayscale and threshold first
gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
_, thresh = cv2.threshold(gray, 127, 255, cv2.THRESH_BINARY)
contours, hierarchy = cv2.findContours(thresh, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

Quick Reference

Contour Retrieval Modes:

• cv2.RETR_EXTERNAL: Retrieves only the extreme outer contours.
• cv2.RETR_LIST: Retrieves all contours without establishing any hierarchical relationships.
• cv2.RETR_TREE: Retrieves all contours and reconstructs a full hierarchy of nested contours.

Contour Approximation Methods:

• cv2.CHAIN_APPROX_NONE: Stores all contour points.
• cv2.CHAIN_APPROX_SIMPLE: Compresses horizontal, vertical, and diagonal segments and leaves only their end points.

Key Takeaways

Always convert images to binary format before using cv2.findContours for accurate detection.

Use cv2.RETR_EXTERNAL to get outer contours or cv2.RETR_TREE for full hierarchy of contours.

cv2.CHAIN_APPROX_SIMPLE reduces memory by compressing contour points without losing shape.

Contours can be drawn with cv2.drawContours to visualize detected shapes.

Check OpenCV version to handle findContours return values correctly.