UE20CS333: Image Processing and Computer Vision 1 Unit I Gowri Srinivasa Department of Computer Science and Engineering Introduction To Digital Image Processing and Computer Vision Slides collated by: Mr. Yashas Kadambi, VIII Semester, Department of CSE, PESU2023 yashasks@pesu.pes.edu With grateful thanks for contribution of slides to: Mr. Deepank Girish, VIII Semester, Department of CSE, PESU2022 Gowri Srinivasa Department of Computer Science and Engineering Teaching Team 2023 Course Instructor TA for Unit 1 TA for Unit 2 Dr. Gowri Srinivasa Email: gsrinivasa@pes.edu Mr. Yashas Kadambi Email: yashasks@pesu.pes.edu Ms. Aanchal Narendran Email: aanchalnarendran@gmail.com TA for Unit 3 TA for Unit 4 TA for Unit 5 Ms. Anushka Hebbar Mr. Harshith Mohan Kumar Email: harshithmohankumar@pesu.pes.edu Email: anushkahebbar@pesu.pes.edu Mr. Utkarsh Gupta Email: utkarsh348@gmail.com Topics Covered In Unit 1 1. 2. 3. 4. 5. 6. Introduction to digital image processing Origins, example fields and various components of DIP Basics of visual perception Image acquisition Sampling and Quantization Relationship between pixels and Review of relevant Linear Algebra Concepts 7. Basics of spatial processing, Negative, log, Power law, Piece wise linear functions 8. Histograms and using histogram statistics for processing 9. Histogram equalization and matching 10. Mechanics of spatial filtering, correlation & convolution Some Prerequisites Basic mathematics required for this course:• Vector and Matrix operations • Set Theory • Probability Typically it is a combination of Linear Algebra, Set Theory and Probability. This will be revised in the course. Definition and basic Operations Digital Image Processing is the computer manipulation of pictures, or images, that have been converted into numeric form Some Basic operations on images are • • • • • • Image Compression Image Warping Contrast Enhancement Blur Removal Feature Extraction Pattern Recognition Need for DIP DIP is a subclass of signal processing specifically concerned with picture images. The aim of doing DIP is:- • Develop methods and applications • to compress images • to provide efficient storage and transmission • To improve image quality for • human perception (subjective) • computer interpretation (objective) Differences between IP, CP, CV, CG Difference between image processing, computational photography, computer vision and computer graphics Image Processing Image to Image Computational Photography Image to Image Computer Vision Image to Model Computer Graphics Model to Image 8 Intersection between Computer Graphics and Computer Vision rendering surface design animation user-interfaces modeling - shape - light - motion - optics - images IP shape estimation motion estimation recognition 2D modeling Computer Graphics Computer Vision Continuum from Image Processing to Computer Vision Image Processing Image-In / Image Out Image Computer Graphics Description In Image Out Low Level Texture mapping Antialiasing Mid Extract attributes Level Edge Detection High Level Noise reduction Contrast enhancement Filtering Computer Vision Segmentation Image In Features out Object recognition Cognitive Functions Scene Description AI Features In Description Out Continuum from Image Processing to Computer Vision The continuum from image processing to computer vision can be broken up into low, mid and high level processes:- Digital image processing focuses on two major tasks – Improvement of pictorial information for human interpretation – Processing of image data for storage, transmission and representation for autonomous machine perception Historical Background Historical Background Wilhelm Conrad Röntgen, born 8 November 1895 • discovery of X-rays – first medical application 1896 • first X-ray image published live experiment demonstrated at PhysikalischMedizinische Gesellschaft Würzburg, Germany • Below image taken on 22.12.1895 - Anna Berthe Röntgen, Hand mit Ringen (hand with rings) Historical Background Early 1920s:- • Bartlane cable picture transmission system • used to transmit newspaper images across the Atlantic • images were coded, sent by telegraph, printed by a special telegraph printer • took about three hours to send an image, first systems supported 5 grey levels Historical Background Image transmitted via Telegraph • using the Bartlane cable picture transmission system • images were transferred by submarine cable between London and New York • printed using a special printer rigged with typefaces simulating halftones Historical Background Mid to late 1920s:• Improvements to the Bartlane system resulted in higher quality images • New reproduction processes based on photographic techniques increased number tones (analog photography) Improved digital image Early 15 tone digital image Historical Background Improved up to an amazing 15 levels of gray Historical Background • When computer vision first started out in the early 1970s, it was viewed as the visual perception component of an ambitious agenda to mimic human intelligence and to endow robots with intelligent behaviour • At the time, it was believed by some of the early pioneers of Artificial Intelligence and Robotics (at places such as MIT, Stanford and CMU) that solving the “visual input” problem would be an easy step along the path to solving more difficult problems such as higher-level reasoning and planning • According to one well-known story, in 1966, Marvin Minsky at MIT asked his undergraduate student Gerald Jay Sussman to “spend the summer linking a camera to a computer and getting the computer to describe what it saw” (Boden 2006, p. 781) Medical Imaging • Emergence of medical imaging • 1979 Nobel Peace Prize for Invention of Computerized Axial Tomography (CAT) • Sir Godfrey Housefield • Professor Allan Cormack Medical Imaging 1980s: the use of digital image processing techniques has exploded and they are now used for all kinds of tasks in all kinds of areas – Image enhancement/restoration – Artistic effects – Medical visualisation – Industrial inspection – Law enforcement – Human Computer Interfaces(HCI) Advanced Techniques • Morphing and visual effects algorithms • JPEG and MPEG compression • Wavelet Transforms Image morph from Michael Jackson Music Video: Black or White Dr. Who Image morphs antonybennison.com Advanced Techniques 2000:• Image based modelling and rendering rely on a set of two-dimensional images of a scene to generate a 3D model and then render some novel views of this scene • Texture synthesis and inpainting • Computational photography • Feature based recognition • MRF inference algorithms • Category recognition learning • Datasets like ImageNet, CIFAR10 and coco available • Start of Video processing Image Storage Formats Some common image formats and their characteristics Format Characteristic jpg/jpeg (Joint Photographic Experts Group) Image compression, supports 8-bit per color (RGB), generational degradation when edited repeatedly tiff (Tagged-Image File Format) Supports 8-bit and 16-bit per color, supports OCR and device-specific color schemes Gif (Graphics Interchange Format) Limited to 256 colors, supports animation png (Portable Network Graphics) 16 million colors (truecolor), good for large images, best suited for editing bmp (Bit Map) Simple, suited for all WINDOWS applications, uncompressed MCQ 1 What are the categories of digital image processing? 1. Image Enhancement 2. Image Classification and Analysis 3. Image Transformation 4. All of the mentioned MCQ 2 Which of the following is the abbreviation of JPEG? 1. Joint Photographic Experts Group 2. Joint Photographs Expansion Group 3. Joint Photographic Expanded Group 4. Joint Photographic Expansion Group THANK YOU Gowri Srinivasa Professor, Department of Computer Science and Engineering Email: gsrinivasa@pes.edu