Computer Vision: Edges and Scale

Computer Vision: Edges and Scale

Edges and Scale From Sandlot Science Todays reading Cipolla & Gee on edge detection (available online) Szeliski 3.4.1 3.4.2 Origin of Edges surface normal discontinuity depth discontinuity surface color discontinuity illumination discontinuity

Edges are caused by a variety of factors Detecting edges Whats an edge? intensity discontinuity (= rapid change) How can we find large changes in intensity? gradient operator seems like the right solution Effects of noise Consider a single row or column of the image Plotting intensity as a function of position gives a signal

Where is the edge? Solution: smooth first Where is the edge? Look for peaks in Associative property of convolution This saves us one operation: Laplacian of Gaussian Consider

Laplacian of Gaussian operator Where is the edge? Zero-crossings of bottom graph 2D edge detection filters Laplacian of Gaussian Gaussian

derivative of Gaussian is the Laplacian operator: The Sobel operator Common approximation of derivative of Gaussian -1 0 1 1

2 1 -2 0 2 0

0 0 -1 0 1 -1 -2 -1 The standard defn. of the Sobel operator omits the 1/8 term

doesnt make a difference for edge detection the 1/8 term is needed to get the right gradient value, however The effect of scale on edge detection larger larger Scale space (Witkin 83) Some times we want many resolutions

Known as a Gaussian Pyramid [Burt and Adelson, 1983] In computer graphics, a mip map [Williams, 1983] A precursor to wavelet transform Gaussian Pyramids have all sorts of applications in computer vision Gaussian pyramid construction filter kernel Repeat Filter (called prefiltering) Subsample (faster approach: filter only of pixels)

Until minimum resolution reached can specify desired number of levels (e.g., 3-level pyramid) The whole pyramid is only 4/3 the size of the original image! Subsampling with Gaussian pre-filtering G 1/8 G 1/4 Gaussian 1/2 Filter the image, then subsample Subsampling with Gaussian pre-filtering

Gaussian 1/2 G 1/4 Filter the image, then subsample G 1/8 Subsampling without pre-filtering 1/2

1/4 (2x zoom) 1/8 (4x zoom) Sampling and the Nyquist rate Aliasing can arise when you sample a continuous signal or image occurs when your sampling rate is not high enough to capture the amount of detail in your image

Can give you the wrong signal/imagean alias formally, the image contains structure at different scales called frequencies in the Fourier domain the sampling rate must be high enough to capture the highest frequency in the image To avoid aliasing: sampling rate 2 * max frequency in the image said another way: two samples per cycle This minimum sampling rate is called the Nyquist rate

Image resampling So far, we considered only power-of-two subsampling What about arbitrary scale reduction? How can we increase the size of the image? d = 1 in this example 1 2 3

4 5 Recall how a digital image is formed It is a discrete point-sampling of a continuous function If we could somehow reconstruct the original function, any new image could be generated, at any resolution and scale Image resampling So far, we considered only power-of-two subsampling What about arbitrary scale reduction? How can we increase the size of the image?

d = 1 in this example 1 2 3 4 5

Recall how a digital image is formed It is a discrete point-sampling of a continuous function If we could somehow reconstruct the original function, any new image could be generated, at any resolution and scale Image resampling So what to do if we dont know Answer: guess an approximation Can be done in a principled way: filtering 1 d = 1 in this

example 1 2 2.5 3 4 Image reconstruction Convert

to a continuous function Reconstruct by cross-correlation: 5 Resampling filters What does the 2D version of this hat function look like? performs linear interpolation

(tent function) performs bilinear interpolation Often implemented without cross-correlation E.g., http://en.wikipedia.org/wiki/Bilinear_interpolation Better filters give better resampled images Bicubic is common choice fit 3rd degree polynomial surface to pixels in neighborhood

Recently Viewed Presentations

  • CSCD 487/587 Human Computer Interface Winter 2013 Lecture

    CSCD 487/587 Human Computer Interface Winter 2013 Lecture

    They are numbered to make it easier to find a particular one (see Table 6.1 for complete set). It should be noted, however, that this classification is for convenience. The interface entries are not mutually exclusive since some products can...
  • Superheroes to the Rescue by Mary Erickson, Ph.D.

    Superheroes to the Rescue by Mary Erickson, Ph.D.

    Superheroes. to the Rescue. by Mary Erickson, Ph.D. and Arizona art teacher, Laurie Eldridge, Ph.D. A Superhero or Superheroine is: A character who has extraordinary or superhuman powers, Is dedicated to protecting the public, (has a strong moral code),
  • International Training Familiarization Course

    International Training Familiarization Course

    Initially your user Name will be ITI and your Instructor Number, (ie ITI10000) Same with your password, it will be ITI and Zip Code (ieITI91706) Once you log on you can change both your user name and password under the...
  • Activity 3.2.1 Fluid Power Applications Exemplar

    Activity 3.2.1 Fluid Power Applications Exemplar

    Compact Utility Loader Fluid Power Applications Project Example Principles of Engineering What Is It? The Powerhouse Prodigy TK-216 compact utility loader with attachments Trencher Auger Ripper Bucket Leveler Used for any number of landscape and trenching applications Power Supply and...
  • Army Career and Alumni Program (ACAP) A Leaders

    Army Career and Alumni Program (ACAP) A Leaders

    MG Sean J. Byrne Commanding General US Army Human Resources Command ACAP Is Congressionally mandated A promise Army makes at enlistment - a program that enables Soldiers to capitalize on their Army experience and skills Something your Soldiers and you...
  • Warm Up # 29

    Warm Up # 29

    Thesis Statement Assessment. 1. On Google Classroom click on the assignment labeled "Analytical Thesis Statement Quiz" 2. Complete the form and hit "submit" This will be graded and placed in the "Assignments" category of your grade.
  • Chapter 3 Software Computer Concepts 2014 3 Chapter

    Chapter 3 Software Computer Concepts 2014 3 Chapter

    030100The two main categories of software are application software and system software. 030200 Location-based software can pinpoint its location by using the device's built-in GPS or by triangulating the distance from nearby cell towers. 030300 Screen readers are an example...
  • The Whole stem - LSMCE Conference

    The Whole stem - LSMCE Conference

    The Whole stem. Building Community Through Programming That Encompasses All STEM Disciplines ... Desire and struggle to collaborate due to schedules, institutional differences, and geography. Enter the symposium ... Riddles 3 and 4. Riddle 4 - casting a shadow. 01010000...