Optical Character Recognition (OCR) MasterClass in Python

Open and view images using the pillow library in python, loading test image.jpg from the same folder with image.open, and inspect the image object's mode rgb and size.

Learn to inspect an opened image with PIL, retrieving the file name, format, mode, and size, including width and height, from an IMG object.

Learn to rotate and resize images in Python using Pillow, rotating by 90 and 180 degrees and resizing to 400x250 and 200x200.

Learn to crop images in Python with Pillow using the image.crop method. Open an image, inspect its size, set left, top, right, and bottom coordinates, and produce a cropped result.

Learn to add text on an image with Pillow in Python by importing Image and ImageDraw, creating a draw object, and using draw.text at specified coordinates with color.

Learn to add padding to an image in Python using Pillow, by setting left, right, top, and bottom values and pasting into a padded image.

Blur images in Python using Pillow with box blur, Gaussian blur, and simple blur methods. Adjust blur intensity with radius and convolutional kernels using image.filter and ImageFilter.

Concatenate two images in Pillow using Python by stacking horizontally or vertically, resizing when needed, and pasting to create a single combined image for OCR workflows.

Learn to save images in python's image library using the save method after rotating or processing, saving intermediate and final results as jpg files in the same directory.

Open an image with OpenCV in Python using cv2.imread, inspect the pixel array as a numpy array, and display the image with matplotlib for OCR preprocessing.

Learn how to invert an image in the OCR masterclass in Python using cv2 bitwise not, turning white-background black-text images into black-background white-text displays.

Learn to binarize a color image to black and white in OpenCV by converting to grayscale and applying thresholding, then save and display the binary image.

Master the erosion and dilation preprocessing steps in optical character recognition with Python, including image inversion, thinning fonts via erosion, and thickening fonts via dilation.

Learn to extract and display text from images in Python using pytesseract and Pillow, loading an image, applying image_to_string, storing the result in a text variable, and printing it.

Detect important words in images and draw bounding boxes around them using cv2 and Python, leveraging image_to_data and dict output to segment text on invoices.

Learn text template matching in an OCR workflow using Python, building a date regex to locate and draw bounding boxes around dates in images.

Detect license plates in images with OpenCV and imutils, convert to grayscale, reduce noise, detect edges, find top contours, crop the plate, and perform OCR with image_to_string in English.

Explore how machine learning enables OCR by training models on labeled data, focusing on supervised learning to classify characters and digits, with practical examples of OCR tasks.

Learn the k-nearest neighbours algorithm, a non-parametric, lazy supervised learner for classification (and regression) using euclidean distance to classify new data in a Python Jupyter notebook.

Train a handwritten digit OCR in python using opencv and k-nearest neighbors by preprocessing an image, splitting into 20×20 blocks, training with digits 0–9, and achieving around 91.64% accuracy.