Note for ECE 533 with Professor Bilgin at UA
Note for ECE 533 with Professor Bilgin at UA
Popular in Course
Popular in Department
This 42 page Class Notes was uploaded by an elite notetaker on Friday February 6, 2015. The Class Notes belongs to a course at University of Arizona taught by a professor in Fall. Since its upload, it has received 19 views.
Reviews for Note for ECE 533 with Professor Bilgin at UA
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 02/06/15
ECEOPTI 533 Digital Image Processing Introduction Ali Bilgin Department of Biomedical Engineering Department of Electrical amp Computer Engineering Fa2010 ECE 533 AB 1 Description of Course This course is designed to provide students with theoretical knowledge and practical experience to analyze and design digital image processing systems The first part of this course covers twodimensional signals and systems We will study extension of key digital signal processing concepts such as sampling ztransforms discrete Fourier transforms and filtering to two dimensions We will then discuss properties of the human visual system and study image enhancement restoration and compression methods Special emphasis will be placed on practical implementation of these methods Fall 2010 ECE 533 AB Course Objectives Upon the completion of this course students are expected to achieve the following objectives 0 Have a fundamental understanding of twodimensional signals and systems 0 Become familiar with image processing methods for enhancement restoration and compression 0 Be able to implement and evaluate the performance of different image processing methods Fall 2010 ECE 533 AB Location and Time TuTh 1100AM 1215PM Civil Engineering Rm 201 Fall 2010 ECE 533 AB Instructor Information Ali Bilgin PhD Office ECE 456H Phone 520 6269414 E mail bilginemailarizonaedu Office hours TuTh 100PM2200PM or by appointment Fall 2010 ECE 533 AB Book liullhllmlm Signal lllllgll Mmmwl Fall 2010 ECE 533 AB llllllil 39 39 iil39i39nl MM Multidimensional Signal Image and Video Processing and Coding John Woods ISBN 012088516 6 published by Elsevier Additional reading will be assigned regularly Course Web Page httpd2arizonaedu Fall 2010 ECE 533 AB Prerequisites ECE 529 Digital Signal Processing ECE 503 Probability and Random Processes for Engineering Applications ECE 529 covers the fundamentals of digital signal processing which are employed extensively in ECEOPTI 533 You will be hopelessly lost in this course if you have not acquired a sound understanding of these principles Fall 2010 ECE 533 AB Administrative Details and Policies Please read the syllabus carefully Information on Attendance participation homework withdrawals academic integrity etc Fall 2010 ECE 533 AB Grading Policy Midterm Exam Final Exam Homework Project 20 20 30 30 See the syllabus for details Fall 2010 ECE 533 AB 10 Programming Assignments Practical results are emphasized in this course Given an image and a processing goal ie color correction sharpening warping etc you will be expected to select and implement an appropriate procedure to achieve that goal Good practical results often depend on an understanding ofthe theory behind the procedures as well as the ability to write software to implement the theory Thus there are significant mathematical and computational components to this course The programming assignments will require knowledge of Matlab as well as C Writing and debugging computer programs will take significant time Fa2010 ECE 533 AB 11 WHAT DOES ECEOPTI 533 COVER ZD signals and systems Sampling in 2 D 2 D z transform 2D discrete Fourier transform 2D discrete cosine transform Subbandwavelet transforms 2D filter design Properties of the human visual system Image enhancement Image restoration Image compression Fall 2010 ECE 533 AB 12 Image Formation Object Fall 2010 ECE 533 AB Lens Image Plane 13 Image Formation Projection through the lens Image of object Fall 2010 ECE 533 AB 14 Image Formation Sensors acquire data Fall 2010 ECE 533 AB Sampling and Quantization 15 Quantization Fall 2010 ECE 533 AB Discrete output Continuous input 16 Quantization Quantization Real image Continuous space Continuous intensity Quantized Continuous space Discrete intensity Fa2010 ECE 533 AB 17 Sampling Sampling Real image Continuous space Continuous intensity Sampled Discrete space Continuous intensity Fa2010 ECE 533 AB 18 Sampling and Quantization Digital Image Sampling 1 Quantization I Real image Sampled amp Quantized Continuous space Discrete space Continuous intensity Discrete intensity Fa2010 ECE 533 AB 19 Digital Image A grid of squares with each square containing a single color 72 63 34 151187115 Each square is called a picture element pel or pixel 70 1 160 Fa2010 ECE 533 AB 20 Color Fall 2010 ECE 533 AB Color images are often created using three intensity maps Each intensity map is projected through a color filter to create a monochrome image The resulting intensity maps are overlaid to create a color image Thus each pixel in a color image can be represented using a vector with three elements 21 Point Processing Original Image 1 Brightness Fall 2010 ECE 533 AB Human Visual System Anatomy of the Eye Anterior chamber 39 bt th r e 0 Most of the focusmg IS actually 333332 333132quot e done by the cornea mg Retina Optic Nerve 0 Objects in the distance are focused on the retina where the rods amp cones are located Pup quot Macu39a ster39o Igigernhlerr Retnal blood region behind vessels the iris Iquot cillary muscle gt Vitreous body II39I a 7 f Conjuctive Myjmlnellet lnc httpenwikipediaorgwikiEye Fall 2010 ECE 533 AB Human Visual System photo receptors 0 The lightsensing retina has a multilayered structure 5 g lgi horizontal bk I I 0 In addition to the rod and cones quotmama amacrine that actually detect the light cells there is a network of cells designed to get the information to the brain for processing cellsaxons 39 Fig 3 Light micrograph of a vertical section through central human retina httpwebvisionmedutaheduimageswvhusectjpeg Fall 2010 ECE 533 AB Human Visual System V cone peak 03 cone density near fovea rods GPE EEC I I mmox ustm In39u 6 40 3 wensJan n n worn m 20301 50mm 30 vu v mm 0 ea mm sccemmcm m degrees estaberg39 1935 Fry 20 Graph to show rod and cone densm39es along the horizontal meridian blue ltlt red lt green Human retina Fig 1 Human retina as seen through an aplhalrnoscope 19992007 by Richard Alan Peters Fall 2010 ECE 533 AB Image Perception Human sensitivity to contrast varies with spatial frequency Fall 2010 ECE 533 AB 26 Color Perception We perceive luminance and Chrominance huesaturation with different resolutions as are red green and blue Original Image Subsampled Red Subsampled Green Subsampled Blue Fa2010 ECE 533 AB 27 Color Perception We perceive luminance and Chrominance huesaturation with different resolutions as are red green and blue Original Image Subsampled Red Subsampled Green Subsampled Blue Fa2010 ECE 533 AB 28 Color Perception We perceive luminance and Chrominance huesaturation with different resolutions as are red green and blue Original Image Subsampled Luminance Subsampled Chrominance Fa2010 ECE 533 AB 29 Image Filtering Input Image Output Image Fa12010ECE533 AB I 30 Image Filtering Fall 2010 ECE 533 AB 31 Image Filtering Fall 2010 ECE 533 AB 32 Image Filtering LPF along columns LPF along rows Fall 2010 EOE 533 AB l39 L1 l 1 L3 7 3339 Image Filtering Fourier Domain Original Image Original Spectrum 2 Rectangwar Window applied to the spectrum Fa2010 ECE 533 AB 34 Image Filtering Fourier Domain Bandpass Filtering Fourier Domain Original Image Bandpass Filter Fitered Image Fall2010 ECE 533 AB 36 Noisy iid Gaussian Fa2010 ECE 533 AB 37 Noisy iid Gaussian Fa2010 ECE 533 AB 38 Noisy iid Gaussian Fa2010 ECE 533 AB 39 Noisy Salt amp Pepper Noisy iid Gaussian Fa2010 ECE 533 AB 4O Image Denoising Pa 2010 ECE 533 AB 41 Original Fall 2010 ECE 533 AB Image Compression JPEGZOOO Compressed Requires only 15 of the original storage 0015 N 42
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'