OpenCV Functions: A Task-Based Python Reference for Computer Vision

A mask is generally a single-channel 8-bit array in which nonzero pixels select output pixels. Related operations are bitwise_or, bitwise_not, split and merge:

b, g, r = cv2.split(image)
merged = cv2.merge([b, g, r])
blue = image[:, :, 0]

Simple NumPy slicing is often clearer for channel access. The full core-array reference is core array operations.

Draw annotations and shapes

cv2.line(image, (10, 10), (200, 100), (0, 255, 0), 2)
cv2.rectangle(image, (50, 50), (200, 150), (255, 0, 0), 2)
cv2.circle(image, (320, 240), 50, (0, 0, 255), -1)
cv2.putText(image, 'Object', (50, 50),
            cv2.FONT_HERSHEY_SIMPLEX, 1, (255, 255, 255), 2)

Coordinates are (x, y), colors are BGR, and a negative thickness usually fills a shape. Text placement uses the baseline rather than a top-left corner. Also useful are polylines, fillPoly, ellipse, arrowedLine and getTextSize. See drawing functions.

Blur and filter images

Choose a filter

blurred = cv2.blur(image, (5, 5))
smoothed = cv2.GaussianBlur(image, (5, 5), 0)
cleaned = cv2.medianBlur(image, 5)
preserved = cv2.bilateralFilter(image, 9, 75, 75)

Gaussian and median kernel sizes are normally positive odd numbers. Filtering is not automatically enhancement: excessive smoothing removes information. See filter reference.

Threshold, segment and clean masks

threshold, Otsu and adaptiveThreshold

gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
_, binary = cv2.threshold(gray, 127, 255, cv2.THRESH_BINARY)
_, otsu = cv2.threshold(gray, 0, 255,
                        cv2.THRESH_BINARY + cv2.THRESH_OTSU)
adaptive = cv2.adaptiveThreshold(
    gray, 255, cv2.ADAPTIVE_THRESH_GAUSSIAN_C,
    cv2.THRESH_BINARY, 11, 2)

threshold returns the threshold actually used and the output image; keep both return values. Otsu works best when the histogram is reasonably bimodal. Adaptive thresholding handles spatially varying illumination; its block size must be odd and greater than one.

Rank #3

Sale

Yilador Webcam Cover 3 Pack, 0.03 inch Ultra Thin Laptop Camera Cover Slide

• Note: Not suitable for MacBooks released after 2023 or devices with a protruding front camera; Not applicable to full-screen or notch-style tempered glass screen protectors; Do not use on the rear camera of the phone.
• 💻 Why Do You Need a Webcam Cover Slide? — Safeguard your privacy by covering your webcam with our reliable webcam cover when not in use. Don't let anyone secretly watch you. Stay protected!
• ✅ Thin & Stylish — Enhance your laptop's functionality and aesthetics with our 0.027" ultra-thin webcam covers. Seamlessly close your laptop while adding a touch of sophistication.
• ✅ Fits Most Devices — Compatible with laptops, phones, tablets, desktops! Keep your privacy intact on Ap/ple, Mac/Book, iPh/one, iP/ad, H/P, L/novo, De/ll, Ac/er, As/us, Sa/msung devices.
• ✅ 365 Days Protection — Our upgraded 3.0 adhesive ensures a strong hold that won't damage your equipment. Experience reliable, long-term privacy protection day in and day out.

$3.99

Buy on Amazon

inRange for color masks

hsv = cv2.cvtColor(image, cv2.COLOR_BGR2HSV)
lower = (35, 50, 50)
upper = (85, 255, 255)
mask = cv2.inRange(hsv, lower, upper)

These values are starting points, not universal settings. The thresholding reference is imgproc miscellaneous functions.

Morphological operations

kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (5, 5))
eroded = cv2.erode(mask, kernel, iterations=1)
dilated = cv2.dilate(mask, kernel, iterations=1)
opened = cv2.morphologyEx(mask, cv2.MORPH_OPEN, kernel)
closed = cv2.morphologyEx(mask, cv2.MORPH_CLOSE, kernel)

Opening removes small isolated foreground regions; closing fills small holes and joins nearby regions. Larger kernels or more iterations can erase small objects or merge objects that should remain separate. Other operations include gradient, top-hat and black-hat. See morphological operations.

Detect edges and analyze contours

cv2.Canny

gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
gray = cv2.GaussianBlur(gray, (5, 5), 0)
edges = cv2.Canny(gray, 50, 150)

The two thresholds control sensitivity and must be tuned for the camera, lighting, resolution and materials. See the Canny tutorial.

findContours and contour measurements

contours, hierarchy = cv2.findContours(
    binary, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

for contour in contours:
    area = cv2.contourArea(contour)
    perimeter = cv2.arcLength(contour, True)
    x, y, w, h = cv2.boundingRect(contour)
    box = cv2.minAreaRect(contour)
    approx = cv2.approxPolyDP(contour, 0.02 * perimeter, True)

Contours usually require a clean binary mask rather than an arbitrary color image. Other useful functions are drawContours, moments, convexHull, isContourConvex, fitEllipse and minEnclosingCircle.

Guard centroid calculations against zero area:

moments = cv2.moments(contour)
if moments['m00'] != 0:
    cx = int(moments['m10'] / moments['m00'])
    cy = int(moments['m01'] / moments['m00'])

Shape-analysis details are in structural analysis.

Rotate, warp and correct perspective

matrix = cv2.getRotationMatrix2D(center, angle, scale)
rotated = cv2.warpAffine(image, matrix, (width, height))

matrix = cv2.getPerspectiveTransform(source_points, destination_points)
warped = cv2.warpPerspective(image, matrix, (output_width, output_height))

Affine transforms use corresponding points and an output size. Perspective correction needs four corresponding corner points. Rotation can crop content unless you enlarge the output canvas. Interpolation and border mode affect the result. Related APIs include getAffineTransform, remap and getOptimalNewCameraMatrix. See geometric transformations.

Measure histograms and improve contrast

histogram = cv2.calcHist([gray], [0], None, [256], [0, 256])
equalized = cv2.equalizeHist(gray)

clahe = cv2.createCLAHE(clipLimit=2.0, tileGridSize=(8, 8))
enhanced = clahe.apply(gray)

equalizeHist is for grayscale images. CLAHE works locally and limits contrast amplification, but either method can amplify noise and cannot recover detail that was never captured. References: histogram functions and the CLAHE tutorial.

Process cameras and video

VideoCapture and the frame loop

cap = cv2.VideoCapture(0)
if not cap.isOpened():
    raise RuntimeError('Could not open camera or video')

while True:
    ok, frame = cap.read()
    if not ok:
        break
    cv2.imshow('Video', frame)
    if cv2.waitKey(1) & 0xFF == ord('q'):
        break

cap.release()
cv2.destroyAllWindows()

Use a filename instead of 0 for a video file. Camera properties are requests, not guarantees; drivers and backends may ignore unsupported width, height or frame-rate settings:

print(cap.get(cv2.CAP_PROP_FRAME_WIDTH))
print(cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
print(cap.get(cv2.CAP_PROP_FPS))

Try another camera index, lower resolution, a different backend or released applications when opening succeeds but reads fail. See VideoCapture and the video-I/O overview.

Rank #4

AboveTEK Portable Laptop Lap Desk w/Retractable Left/Right Mouse Pad Tray, Non-Slip Heat Shield Tablet Notebook Computer Stand Table w/Sturdy Stable Work Surface for Bed Sofa Couch or Travel

• Anti-Slip Surface - Transform your laptop into a mobile workstation with the AboveTEK portable laptop lap desk. The anti-slip surface provides a strong grip for laptops up to 15.6 inches(Diagonal), while the double rubber strip on the bottom ensures a stable display or typing experience on your lap, couch, or bed.
• Retractable Mouse Pad - Retractable laptop mouse pad extends on both directions for the left/right handed with elevation along the edges for stopping mouse from falling off. The size of laptop tray is 14" X 9.7" and the size of mouse pad is 7.4" X 6.1".
• Effective Heat Shield - The effective heat shield made of sturdy and thick material protects your laptop from overheating. Prioritizes your comfort and safety, an ideal lap pad or board for working anywhere.
• EASY to Carry and Store - With an ergonomic and simplistic design, the lap desk is portable to store in a backpack. Only 15" in size, 2.2 lb of weight and with slim 0.6 inch thickness, it is ready to be easily carried around.
• Widely Applicable - The smooth platform accommodates laptops and tablets up to 15.6 inches(Diagonal), making it a versatile accessory and one of the best gifts for mom, dad, students and professionals. Perfect for use as a laptop bed tray or tablet holder anywhere at home, library, or park.

$25.99

Buy on Amazon

VideoWriter

fourcc = cv2.VideoWriter_fourcc(*'mp4v')
writer = cv2.VideoWriter('output.mp4', fourcc, 30.0, (width, height))
if not writer.isOpened():
    raise RuntimeError('Video writer could not be opened')

writer.write(frame)
writer.release()

Frame dimensions must exactly match the writer dimensions. Codec and container support depends on platform backends and installed codecs, so a valid-looking writer does not guarantee a playable file. Always call release. See VideoWriter.

Features, descriptors and matching

gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
orb = cv2.ORB_create()
keypoints, descriptors = orb.detectAndCompute(gray, None)

matcher = cv2.BFMatcher(cv2.NORM_HAMMING, crossCheck=True)
matches = matcher.match(descriptors_a, descriptors_b)

ORB uses binary descriptors and is often chosen for speed. SIFT_create is another option that is generally more tolerant of scale and rotation, with different performance and deployment considerations. Other APIs include BFMatcher, FlannBasedMatcher, drawKeypoints and drawMatches. Matching local features is not semantic object detection and can fail with viewpoint changes, blur, occlusion or repetitive textures. See features2d.

Calibrate cameras and estimate 3D geometry

Calibration is a dataset-and-validation process, not a single call. A typical workflow is:

1. Capture many views of a target with known geometry, such as a chessboard.
2. Detect corners with findChessboardCorners and refine them with cornerSubPix.
3. Build corresponding 3D object points and 2D image points.
4. Call calibrateCamera, then validate on images not used for calibration.

Important APIs include undistort, getOptimalNewCameraMatrix, solvePnP, projectPoints, stereoCalibrate, stereoRectify and reprojectImageTo3D. Documentation: calibration tutorial and calib3d reference. OpenCV 5 reorganizes portions of former calib3d functionality; check the installed version's generated documentation.

Classical detectors and QR codes

cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
objects = cascade.detectMultiScale(
    gray, scaleFactor=1.1, minNeighbors=5)

CascadeClassifier, HOGDescriptor and QRCodeDetector support constrained detection tasks. Haar cascades are not equivalent to modern deep-learning detectors: pose, lighting, occlusion and domain changes can reduce robustness. Barcode and ArUco APIs depend on the installed build and modules. See object detection.

Run trained models with the DNN module

net = cv2.dnn.readNetFromONNX('model.onnx')
blob = cv2.dnn.blobFromImage(
    image, scalefactor=1 / 255.0, size=(640, 640),
    swapRB=True, crop=False)
net.setInput(blob)
output = net.forward()

Other entry points include readNet, blobFromImages, getPerfProfile and backend/target configuration methods. An .onnx suffix alone does not guarantee compatibility. Preprocessing must match training: input size, scaling, mean subtraction, channel order and letterboxing or cropping. Raw output normally needs decoding, confidence filtering and non-maximum suppression. GPU acceleration depends on the OpenCV build and available backend. See the DNN module and DNN tutorials.

Track motion and foreground objects

subtractor = cv2.createBackgroundSubtractorMOG2()
mask = subtractor.apply(frame)

Optical-flow functions include calcOpticalFlowPyrLK and calcOpticalFlowFarneback; another background model is createBackgroundSubtractorKNN. Background subtraction assumes a reasonably stable camera and scene. Shadows, illumination changes, vibration and moving backgrounds produce false positives. A tracker can drift or lose an object and is not the same as a detector. See video analysis.

Specialized photography and stitching functions

For restoration and compositing, consider inpaint, fastNlMeansDenoising, detailEnhance, stylization and seamlessClone. Panorama workflows use the stitching APIs exposed by the installed version; older examples may show createStitcher while newer builds can differ. Consult the photo module and stitching module.

Best Value

Sale

LAPGEAR Home Office Lap Desk – Pink, Fits 15.6” Laptops

• Spacious Design: Measuring 21.1" wide and 12" deep, our lap desk comfortably fits most laptops up to 15.6". Extra room for accessories ensures convenience.
• Enhanced Functionality: Packed with handy features, including a 5x9" precision tracking mouse pad and a built-in phone slot for seamless work or video calls. Plus, enjoy laptop support with the integrated device ledge.
• Cool Comfort: Enjoy a stable surface with our lap desk's dual bolster cushion, designed for comfort and airflow, keeping your lap cool during extended use.
• Durable Surface: Work with confidence on our lap desk's solid surface, featuring a blush pink color, ensuring optimal air circulation to prevent your laptop from overheating.
• On-the-Go Convenience: With an integrated handle and lightweight design (2.14 lbs), our lap desk is portable for travel or moving around the house, offering flexibility in any space.

$36.59

Buy on Amazon

A complete teaching pipeline

import cv2

image = cv2.imread('input.jpg')
if image is None:
    raise FileNotFoundError('input.jpg could not be read')

gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
blurred = cv2.GaussianBlur(gray, (5, 5), 0)
edges = cv2.Canny(blurred, 50, 150)

contours, _ = cv2.findContours(
    edges, cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)

output = image.copy()
for contour in contours:
    if cv2.contourArea(contour) < 100:
        continue
    x, y, w, h = cv2.boundingRect(contour)
    cv2.rectangle(output, (x, y), (x + w, y + h), (0, 255, 0), 2)

if not cv2.imwrite('output.jpg', output):
    raise IOError('output.jpg could not be written')

This demonstrates a common order—load, grayscale, smooth, detect edges, find contours and annotate—but it is not a reliable object detector. Canny edges can create fragmented or duplicate outlines, and contour boxes do not provide semantic labels.

Quick function lookup

Troubleshoot the failures that appear most often

imread returns None

from pathlib import Path
path = Path('input.jpg')
print(path.resolve(), path.exists())
image = cv2.imread(str(path))

Check the working directory, spelling, permissions, file integrity and supported format.

Colors are wrong

Convert BGR to RGB before passing an image to an RGB-oriented library:

rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)

imshow freezes or crashes

Use waitKey and destroyAllWindows in desktop applications. In headless environments, remove GUI calls and write files or use another display mechanism.

Contours are noisy

Improve the binary input: grayscale conversion, selective blur, thresholding or segmentation, morphology, then filtering by area, aspect ratio, solidity or hierarchy.

The camera opens but frames fail

Check both isOpened() and the Boolean returned by read(). Try another index, lower resolution or frame rate, a backend-specific API, and verify operating-system camera permissions.

The video file is empty or unplayable

Confirm that the writer opened, frame dimensions match exactly, the codec is supported, the extension matches the container and release() is called.

DNN output is incorrect

Recheck input dimensions, BGR/RGB order, scaling, mean subtraction, letterboxing, output decoding, confidence thresholds, non-maximum suppression and model/backend compatibility.

Processing is too slow

• Resize frames or process every second or third frame.
• Restrict work to a region of interest.
• Avoid unnecessary array copies and use vectorized NumPy operations.
• Batch DNN inputs where appropriate.
• Measure stages with time.perf_counter() before changing backends or models.

When OpenCV is enough—and when it is not

Use OpenCV alone for local image and video manipulation, deterministic filtering, geometric correction, classical segmentation, camera access and lightweight feature workflows. Add a trained model with OpenCV DNN, ONNX Runtime, PyTorch or another inference stack when you need robust semantic classification, detection or segmentation across changing scenes. OpenCV supplies inference utilities; it does not replace dataset labeling, model training, evaluation, monitoring or production operations.

Choose a managed vision platform only when hosted scaling, annotation, model lifecycle tooling or pre-trained APIs justify their cost, latency, privacy and vendor-lock-in trade-offs. Check the license for the exact OpenCV package, contrib module, model weights, codec and deployment service before shipping a product.

Quick Recap

SaleBestseller No. 1

LAPGEAR Home Office Pro Lap Desk - Black Carbon, Fits 15.6” Laptops

$41.99

Bestseller No. 2

Anker USB C Hub, 5-in-1 USBC to HDMI Splitter with 4K Display

$24.99

SaleBestseller No. 3

Yilador Webcam Cover 3 Pack, 0.03 inch Ultra Thin Laptop Camera Cover Slide

$3.99

Bestseller No. 4

AboveTEK Portable Laptop Lap Desk w/Retractable Left/Right Mouse Pad Tray, Non-Slip Heat Shield Tablet Notebook Computer Stand Table w/Sturdy Stable Work Surface for Bed Sofa Couch or Travel

$25.99

SaleBestseller No. 5

LAPGEAR Home Office Lap Desk – Pink, Fits 15.6” Laptops

$36.59

Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.