Introduction to Image Processing Grass Sky Tree ? ? Introduction A picture is worth more than a thousand words
Aims and Objectives A general introduction to the common techniques of image processing and its relations with Computer Vision and Computer Graphics To review and understand the principal approaches used, which provide as the basis for further study of the related fields To offer practical experience in writing programs that manipulate images, using examples in Java
Module Information Prerequisites: G51MCS, G51PRG or equivalent Course Structure –Lecture: 2 hours/week –Practical: 1 hour/week (starting week 3/4) Assessment –Programming Assignment (40%) –1-hour exam (60%) Handouts OR /G52IIP/ Reference Books –Digital Image Processing, 3 rd Edition by Rafael Gonzalez & Richard Woods –Digital Image Processing Using Java by Nick Efford
Image Processing, Vision, Graphics In image processing we do things like –removing noise from images –finding edges and features in images –generally changing 2D images to other 2D images In computer vision we do things like –finding moving objects in a scene –recognising objects from a database –build abstract models of the world from images –generally extract information about 3D world from 2D images In computer graphics we do things like –creating a 3D model, with realistic shape, colour, texture, and project it to 2D for viewing –animate the 3D model –creating a virtual world and animate the objects in it The link between imaging and graphics –image based modelling
Computer Graphics (Courtesy of Michael Cohen) Image Output Model Synthetic Camera
Real Scene Computer Vision Real Cameras Model Output (Courtesy of Michael Cohen)
Vision and Graphics Combined Model Real Scene Real Cameras Image Output Synthetic Camera (Courtesy of Michael Cohen)
Image Processing (1) Image Enhancement Image Restoration
Image Processing (2) Image Compression
Computer Vision: Recognition
Another Useful Paradigm Three Processing Levels: Low-level process: –primitive operations like noise reduction, contrast enhancement, image sharpening… –input: image output: image Mid-level process: –tasks like segmentation, representation, description –input: image output: attributes extracted from images High-level process –“making sense” of an ensemble of recognized objects –input: image (sequence) output: interpretation Digital Image Processing Computer Vision
Proposed Contents Fundamentals Basic Image Manipulation Noise and Spatial Techniques Point, Line & Edge Detection Image Segmentation & Representation Basic Spectral Techniques Others if time permits
Words of Caution Lecture Contents slightly mathematical + practical attendance required to understand lecture materials not difficult but leaving it until the last minute can be disastrous further readings are required if you intend to undertake DIP related projects Reference Texts read them if you need more detailed explanations than what is found in the lecture slides Tips for Passing this Module coursework must be submitted practise by solving the past exam questions show detailed steps of how the final answers are derived in your answer scripts
Light and Colour The spectrum of electromagnetic waves:
Radio Images kneespinehead visible infraredradio Radio Band − medicine and astronomy
Radar Images mountains in southeast Tibet Microwave Band
Thermal Images human body disperses heat (red pixels) different colours indicate varying temperatures Infrared Band
Remote Sensing hurricane Andrew taken by NOAA GEOS America at night (Nov. 27, 2000) V isible and Infrared Bands − weather and environmental observations
Light Microscopy taxol (250 ) cholesterol (40 )microprocessor (60 ) V isible Band − pharmaceuticals and microinspection
Fluorescence Microscopy normal cornsmut corn U ltraviolet Band
X-Ray Imaging chesthead
Gamma-Ray Imaging positron emission tomography Cygnus Loop in the constellation of Cygnus
Other Non-Electromagnetic Imaging Modalities Acoustic Imaging translate “ sound waves ” into image signals Electron Microscopy shine a beam of electrons through a specimen Synthetic Images in Computer Graphics computer generated (non-existent in the real world)
Application Areas Space cosmic radiation, images from Hubble space telescope and interplanetary probe Medicine multitude of diagnostic medical images Remote Sensing and GIS terrain classification and meteorology Industrial Inspection replace human operators with machines Security and Law Enforcement surveillance and biometrics Human Computer Interaction more natural interface using face, gesture recognition
Acknowlegements Slides are modified based on the original slide set from Dr Li Bai, The University of Nottingham, Jubilee Campus plus adoptions from the following sources: Digital Image Processing book by Gonzalez and Woods robots.stanford.edu/cs223b07/notes/CS223B-L1- Intro.ppt nia/prez/TEI/bebis/CV_Overview.ppt