Learn Computer Vision with OpenCV Library using Python

install python 3 and OpenCV 3 on macOS using brew, update the package manager, install OpenCV 3 with its dependencies, and force link to complete setup.

Install Anaconda for Windows and download OpenCV for Windows, then extract to the C drive. Copy the file into site-packages, set the path, and import cv2 in Python.

Import numpy and OpenCV, read an image from the working directory, display it with waitKey, and explore grayscale, color, and alpha channel options by changing function arguments.

Define image dimensions and create a blank black image, then use OpenCV drawing functions in Python to render various shapes, following along with the code.

Create a 500 by 500 image with red, green and blue channels by initializing a zero array of eight-bit integers, then draw a rectangle with specified coordinates and color.

Create a 500 by 500 image with red, green, blue channels, fill an array with zeros, and draw a line from a start point to an end point in color.

Create a 500 by 500 image, define a three-channel rgb array, and draw a purple circle at the center with radius 50, then try different radii and colors.

Create a 500x500 image with three color channels in Python using OpenCV, then write text on it with putText and font FONTES duplex, white color, and thickness.

Combine a line, text, circle, and rectangle to create a single image using OpenCV with Python. The lecture guides placing shapes with coordinates inside the rectangle for a cohesive composition.

Shift images with OpenCV in Python by applying a translation matrix to move an image 150 pixels right and 70 down, then invert the shift with negative values.

Learn to rotate images with OpenCV in Python by defining the image center, choosing an angle, constructing the rotation matrix, and applying the rotation at the desired scale.

Explore image thresholding in grayscale using OpenCV with Python, setting a threshold value to convert pixels below to black and above to white, and learning inverse threshold.

Learn how gaussian blur uses a 7 by 7 matrix and sigma to weight neighboring pixels, producing a softer image than simple averaging.

Apply median blur to remove salt-and-pepper noise from images with a 3x3 kernel, replacing the central pixel by the median of its neighbors, and compare before-and-after results.

Explore how bilateral filtering preserves edges while smoothing images, using diameter, color, and space parameters, and compare its slower performance and noisier results against other filters.

Explore feature detection with the canny edge detector in OpenCV for Python, load an image, tune two thresholds, and reveal white edges on a black background.

learn to read and stream a video using opencv in python by opening a capture, looping while it is open to read frames, and breaking to end, then destroying windows.

Learn to save a video in a different format by configuring the codec, frames per second, and frame size (720 by 480) and writing frames to a file.

Learn face detection in images using a pre-trained cascade classifier found in OpenCV, and build a custom classifier by gathering many positive and negative images and extracting features.

Enable real-time face detection by switching from image loading to webcam video capture and streaming frames with the existing OpenCV Python code. Watch the loop detecting faces in real time.

Learn to perform image face detection with OpenCV's cascade classifier in Python, loading a pre-trained classifier, adjusting scale factor and minNeighbors, drawing rectangles around detected faces, and counting faces.

Apply existing face detection code to real-time webcam streams with OpenCV in Python, capturing video frames, processing each frame, and printing when a face is detected.