TDI2131 Digital Image Processing

Transcription

1 TDI2131 Digital Image Processing Introduction to Image Processing Lecture 1 John See Faculty of Information Technology Multimedia University Some portions of content adapted from Zhu Liu, AT&T Labs 1

2 Lecture Outline Course Information Introduction & Overview Applications of Image Processing Fundamental Image Processing Operations More Applications... 2

3 Course Instructor John See Office: BR Tel (o): Consultation Hours: Wednesdays, 2-6pm All other times by appointment 3

4 Textbook Digital Image Processing (3 rd Edition) Gonzalez & Woods The version in the bookstore has a purple cover instead of red 4

5 Other References Introduction to Digital Image Processing with MATLAB, A. McAndrew (2004) Fundamentals of Digital Image Processing, A.K. Jain (1990) Course notes from other universities available online 5

6 Lecture & Tutorial Lectures: Every Thursday 4-6pm, CR2003 Tutorials: AR 2003 (GVGD Lab) Every Friday, 10am-12pm Theory & MATLAB exercises 6

7 Grading Assignments 30% Assignment 1 (8%) Assignment 2 (10%) Assignment 3 (12%) Term Test 10% Final Exam 60% 7

8 Coursework Term Test: in Week 12 (most likely) Assignments (using Matlab) Assignment 1 (8%) X-ray Enhancement due Week 7 Assignment 2 (10%) Fingerprint Feature Extraction due Week 11 Assignment 3 (12%) Automated Color-based Face Detection due Week 14 8

9 Housekeeping Attendance will be taken in both lecture and tutorial. Signing for someone other than yourself is prohibited. Plagiarism (especially in assignments) is an offence. You can be failed or suspended academically if found to have plagiarised others. Participation in class is highly encouraged. Some bonus marks may be awarded to you discretely. Syllabus is non-exhaustive. There is always much more topics/areas that may not be covered within the limits of this course, so self-exploration is highly encouraged! 9

10 Overview Early days of computing data was numerical Later, textual data became more common Today, many other forms of data: voice, speech, images, video, web, wireless packets, etc. Each of these types of data are signals. Loosely defined, a signal is a function that conveys information. 10

11 Relationship of Signal Processing to Other Fields People have tried to send or receive signals through electronic media telegraphs, telephones, television, radar, etc. --- signals affected by the system used to acquire, transmit, or process them. Systems can be imperfect and introduce noise, distortion, or other artifacts Finding a way to correct them is fundamental in signal processing 11

12 Relationship of Signal Processing to Other Fields To send specific signals/messages to others, information content is introduced into the signal and hopefully, we can extract them later! Where do we find these signals? Signals encoded from natural phonemena (audio signals, images from photographs, scenes from video footage) Signals created synthetically, man-made (speech generation, computer graphics) 12

13 Concerned Fields in Signal Processing Digital Communication (Wired& Wireless) Data Compression Speech Synthesis & Recognition Computer Graphics Image Processing (sometimes with video processing) Computer Vision 13

14 Fields that deal with Images Computer Graphics: Creation of images synthetically Image Processing: Enhancment or manipulation of the image the result of which is usually another image Computer Vision: Analysis and understanding of image content Video Processing (new!): Similar with image processing, but processing of multiple images/frames. Combined with computer vision, end result is normally extracted information 14

15 3 Principal Uses of Image Processing Improvement of pictorial information for human interpretation Compression of image data for storage and transmission Processing of image data for autonomous machine perception to enable object representation, detection, classification and tracking 15

16 Categorisation by Image Sources Radiation from Electromagnetic Spectrum Acoustic Ultrasonic Electronic (in the form of electron beams used in electron microscopy) Computer (synthetic images used for modeling and visualization) 16

17 What are some applications that make use of Image Processing? 17

18 Typical Areas of Application Television Signal Processing Satellite Image Processing / Remote Sensing Medical Image Processing Robotics Visual Communications Law Enforcement Automatic Visual Inspection for Manufactured Goods Etc. 18

19 Television Signal Processing Image brightness, contrast, color hue adjustment Video compression for efficient delivery and storage Conversion among different video formats QVGA <-> VGA <-> XVGA SDTV <-> HDTV NTSC <-> PAL 19

20 Medical Image Processing Images are acquired to get information about Anatomy and Physiology of a patient How to reconstruct the image from captured data How to process/analyze the image to help diagnosis/treatment? Ultra Sound (US) Magnetic Resonance Imaging (MRI) Positron Emission Tomography (PET) Computer Tomography (CT) X-Rays 20

21 Visual Communication Videophone Tele-conferencing Tele-shopping How to compress the video to reduce bandwidth/storage requirements? How to conceal artifacts due to transmission losses? 21

22 Law Enforcement Biometric Identification / Verification Fingerprint Face Iris How to extract features that can be used to differentiate among different images? 22

23 Law Enforcement Paper currency or cheque fraud Automated counting or reading of serial number for tracking and identifying bills Automated license plate reading 23

24 Robot Control Automatic Maneuvering Unmanned Operations Autonomous Vehicle Driving How to detect and track target? How to avoid obstacles? Mars Rover 24

25 Automated Visual Inspection Manufactured Goods Circuit board missing parts Pill container missing pills Bottles filled up levels Bubbles in clear-plastic product detect unacceptable air pockets Cereal inspection for color, presence of burnt flake Image of replacement lens for human eye inspection of damaged implants 25

26 Satellite Image Processing Remote sensing Climate Geology Land resource Flood monitor How to enhance the image to facilitate interpretation? How to analyze the image to detect certain phenomena? New York (from Landast-5 TM) 26

27 Remote Sensing Weather Observation and Prediction Multispectral image of Hurricane Andrew from satellites using sensors in the visible and infrared bands 27

28 Acoustic Imaging Cross-sectional image of a seismic model. The arrow points to a hydrocarbon (oil and/or gas) trap (bright spots) 28

29 Components in Digital Image Processing We will deal mainly with most of the light green boxes. Yellow boxes belong to computer vision and pattern recognition 29

30 Image Acquisition Camera Consist of 2 parts Lens: Collects appropriate type of radiation emitted from object of interest and forms and image of the real object Semiconductor device: Charged-coupled device (CCD) which converts the irradiance at image plane into an electrical signal 30

31 Image Acquisition Framegrabber Needs circuits to digitize the electrical signal from the imaging sensor to store the image in the memory (RAM) of the computer. 31

32 3 Levels of Image Processing Low-level Processing: input & output are images Primitive operations such as image preprocessing to reduce noise, contrast enhancement and image sharpening and smoothing. Mid-level Processing: input may be images, output are attributes extracted from those images Segmentation, description of objects, classification of individual objects High-level Processing Image analysis and understanding of content, representation and recognition of extracted patterns from images 32

33 Basic Image Processing Operations Simple Point Processing Image Enhancement Image Restoration Noise Reduction Colour Image Processing Image Segmentation Morphological Image Processing Etc. 33

34 Simple Point Processing 34

35 Image Enhancement To bring out details that are obscured, or to highlight certain features of interest in an image 35

36 Image Enhancement Negative transformation of a digital mamogram. Note that the cancerous region (dark spot) in the right image is enhanced. 36

37 Image Restoration Improving appearance of an image. Tend to be based on mathematical or probabilistic models of image degradation 37

38 Image Restoration: Noise Reduction Improving appearance of an image. Tend to be based on mathematical or probabilistic models of image degradation 38

39 Colour Image Processing Colour is a powerful descriptor that often simplifies object identification and extraction from a scene. Human can discern thousands of colour shades and intensities, compared to about only two dozen shades of gray. 39

40 Image Segmentation Attempts to separate certain objects of interest from the image background or other objects One of the most difficult tasks in DIP! Output of the segmentation stage is raw pixel data, constituting either the boundary of a region or all the points in the region. 40

41 Image Segmentation Edge Detection Colour Region Segmentation Extraction of settlement area from aerial imagery Ground replacement due to earthquake in California,

42 Wavelets and Multi-resolution Processing Foundation of representing images in various degrees of resolution. Used in image data compression and pyramidal representation (images are subdivided successively into smaller regions). 42

43 Image Compression Reducing storage required to save an image or the bandwidth required to transmit it. JPEG, JPEG2000, JBIG2 43

44 Morphological Image Processing Mathematical morphological operations Tools for extracting image components that are useful in the representation and description of shapes. 44

45 Image Morphing (Metamorphosis) Transformation of one digital image to another. Special visual effect in the entertainment industry! 45

46 Image Stitching (Mosaics) Blend together overlapping images to produce a panoramic stitched image = 46

47 And now...what do you see? 47

48 We are smarter than computers! How do you recognize the banana? How can you get a computer to do that? 48

49 Representation and Description Representation make a decision on how the extracted data should be represented 49

50 Recognition and Interpretation Recognition the process that assigns a label to an object based on the information provided by its descriptors. Interpretation assigning meaning to an ensemble of recognized objects. 50

51 Knowledge Base A problem domain detailing regions of an image where the information of interest is known to be located Help to limit search, pinpoint information details to be used for further processing. 51

52 High-level Processing The two last stages in the chart (Representation & Recognition) are often categorized as high-level processing and usually belong to the area of Computer Vision / Pattern Recognition. Depending on the usage or application, the earlier stages can be used to prepare images for high-level processing. Let's see what applications that considered high-level processing... 52

53 Fingerprint Recognition 53

54 Content-based Image Retrieval Query image 54

55 Face Detection Final Year Project 2006 FACEFIND, Nusirwan & Chong 55

56 Face Detection Is it harder to detect faces in a large group of people? 56

57 Tracking and Counting People 57

58 Readings Digital Image Processing (3rd Edition), Gonzalez & Woods, Chapter 1 MATLAB Getting Started Guide from Mathworks More tutorials to get you started (downloadable from my website links): MATLAB Primer Concise but extensive introduction to Matlab Two basic introductory tutorials by Gerald Recktenwald very easy to understand 58