
Explore image compression, separating redundant and irrelevant information to reduce size through lossless and lossy methods, balancing memory use, quality, and reversibility for text and multimedia.
Explore lossless compression with run length encoding (RLE) in image processing, following a five-step workflow and a hands-on example that compares pre- and post-compression sizes.
Learn arithmetic coding for lossless compression by encoding symbol sequences through iterative interval refinement and retrieving the code word from the final interval using max-min times probability plus base.
Learn how Huffman coding applies to image pixels for lossless compression by calculating frequencies and probabilities, building a Huffman tree, assigning codes, and evaluating entropy, average length, and efficiency.
Explore lossy compression for images and videos using jpeg and mpeg. Learn the jpeg pipeline: 8x8 blocks, ycbcr, dct, quantization, zigzag, and dc encoding; plus mpeg's mp4 and mp3 relevance.
Explore image morphology through dilation and erosion, showing how a structuring element adds boundary pixels, with padding, rules, and practical dilation steps.
Explore the hit or miss transformation, a morphological operator that detects local patterns by matching binary image patterns to a structuring element, producing hits or misses and combining results.
Explore the basics of image classification, including binary and multiclass tasks, and compare algorithms such as logistic regression, SVM, and CNN with a hands-on MNIST implementation.
Explore how convolutional neural networks classify images and compare them to support vector machines, implementing CNNs with Keras to distinguish cats and dogs from data preparation to evaluation.
Explore five core tools for image processing and computer vision, including Matlab, Octave, OpenCV, scikit-image, and Scilab, and learn how they support algorithms, debugging, and rapid testing.
Explore how medical image processing enables brain tumor segmentation, automated blood cell counting, kidney segmentation with deep learning, and lung cancer detection using neural networks.
Apply a cnn-based approach to detect lung cancer from histopathological images, balancing an imbalanced kaggle dataset via downsampling and comparing pretrained architectures (vgg16, inceptionv3, resnet) with training from scratch.
Image processing is a method to perform some operations on an image, in order to get an enhanced image or to extract some useful information from it. Well, don't be overwhelmed by all those technical terms, we will help you with the fundamentals. So now are you interested to know about the field of Image Processing? Then this self-paced course is for you!
This course has been designed such that we can share our knowledge and help you learn complex theory, techniques and concepts in a simple way. It is a perfect match for all those self-taught students out there!
We will walk you into the World of Image Processing. With every tutorial you will develop new skills and improve your understanding of this field. While preparing this course special care is taken that the concepts are presented in fun and exciting way but at the same time, we dive deep into Image Processing. Using the strategies, lessons, and exercises in this course, you will learn all that is necessary to be a master in Image Processing. Here is a list of few of the topics we will be learning:
· What is Image Compression
· Lossless compression
· Lossy compression
· Lossy algorithms
· Lossless algorithms
· Blocks of Machine Learning
· Convolutional Neural Network
· Applications of Medical Image Processing
· Live project demonstration