Summary: Computer Vision

Read the summary and the most important questions on Computer Vision

• 1 Spatial and frequently domain processing

  This is a preview. There are 20 more flashcards available for chapter 1
  Show more cards here

• Discrete (Fast) Fourier Transformation

  Applies cos+/-sin basis functions at different frequences, while the weight coefficients represent the amplitude/magnitude of a particualr frequency

• Discrete cosine transformation (jpeg compression)

  Applies only a cosine basis fucntions for decomposition

• Oriented / steerable pyramid

  Applying a filter bank to levels of pyramid

• 2 Gradient domain processing

  This is a preview. There are 24 more flashcards available for chapter 2
  Show more cards here

• Why are edges intereting

  • Object bounderies
  • reflectance changes
  • detect shadows
  • analyze textures
  • analyze surface orientation
    

• Shot noise (photon noise)

  Through variations of photons sensed, Poisson distribution, noise level depends on signal

• Salt-and-pepper noise

  Though A/D conversion or dead pixels, dark pixels in bright regions and bright pixels in dark regeions, mostly static, random distribution

• Gradients in x , y directions of an image forms a

  Vector field

• 3 segmentation and local features

  This is a preview. There are 15 more flashcards available for chapter 3
  Show more cards here

• Fast technique then scale-invariant feature transform

  Speed-Up Robust Features (SURF) or FERNS

• 4 Basics of cameras

  This is a preview. There are 19 more flashcards available for chapter 4
  Show more cards here

• Thick lenses - abberrations

  • Light rays don't intersect in a single point
  • Monochromatic and chromatic aberrations exists
  • minimized by aliging simple lensens with well choses shapes and refraction indices

• Thick lenses - distortion

  • Different areas of the lens have slightly different focal lengths
  • pincusion and barrel distortion