23
Edge detection Goal: Identify sudden changes (discontinuities) in an image • This is where most information in an image is coded Example: line drawings

Edge detection

  • Upload
    erling

  • View
    16

  • Download
    2

Embed Size (px)

DESCRIPTION

Goal: Identify sudden changes (discontinuities) in an image This is where most information in an image is coded Example: line drawings. Edge detection. What causes an edge?. Depth discontinuity Surface orientation discontinuity Changes in surface properties - PowerPoint PPT Presentation

Citation preview

Page 1: Edge detection

Edge detection

• Goal: Identify sudden changes (discontinuities) in an image

• This is where most information in an image is coded

• Example: line drawings

Page 2: Edge detection

What causes an edge?

• Depth discontinuity

• Surface orientation discontinuity

• Changes in surface properties

• Light discontinuities (e.g. shadows)

Page 3: Edge detection

ScaleIncreased scale:• Eliminates noisy edges• Makes edges smoother and thicker• Removes fine details

Page 4: Edge detection

Suppression of non-maxima:• Choose the local maximum point along a perpendicular cross section of the edge.

Page 5: Edge detection

Example: Suppression of non-maxima

courtesy of G. Loy

Original image Gradient magnitudeNon-maxima suppressed

Page 6: Edge detection

Differentiation with a Gaussian filter

Page 7: Edge detection

Example: Canny Edge Detection

Using Matlab with default thresholds

Page 8: Edge detection

Corner Detector

• Compute x- and y-derivatives with a Gaussian filter

• Form the orientation tensor M for every pixel

• Compute the product of eigen-values, i.e., the determinant of M

• If both eigenvalues large (product is a local maximum), then it is a corner!

Page 9: Edge detection

Gradient directions

Page 10: Edge detection

Blow-up of gradient directions

Page 11: Edge detection

Corners can be detected where the product of the ellipse axes are local maxima

Page 12: Edge detection

Fast (bottom-up) - some methods scale

Page 13: Edge detection

Fast (bottom-up) - some methods scale

Page 14: Edge detection

Fast (bottom-up) - some don’t

Page 15: Edge detection

Task: Image Retrieval• Oxford Building Data (Philbin et al. CVPR’07)

Query

Page 16: Edge detection

Task: Image Retrieval• Oxford Building Data (Philbin et al. CVPR’07)

Query

Page 17: Edge detection

Task: Image Retrieval• Oxford Building Data (Philbin et al. CVPR’07)

Good

mat

ch

Query

Page 18: Edge detection

Task: Image Retrieval• Oxford Building Data (Philbin et al. CVPR’07)

Good

mat

ch

Matched?

Query

Page 19: Edge detection

Task: Image Retrieval• Oxford Building Data (Philbin et al. CVPR’07)

Good

mat

ch

Matched?

Query

Page 20: Edge detection

Baseline System – Bags of Words

• Interest point detection (position, scale, orientation)

- Differences of Gaussian/Harris

Page 21: Edge detection

Baseline System – Bags of Words

• Interest point detection

- Differences of Gaussian/Harris

• Feature extraction (feature vector e g R^128)

- SIFT/SURF/DAISY

Page 22: Edge detection

Baseline System – Bags of Words

• Interest point detection

- Differences of Gaussian/Harris

• Feature extraction

- SIFT/SURF/DAISY

• Generating vocabularies – quantization- hierarchical k-means (Nister, Stewenius CVPR’06)

- approximate k-means (Philbin et al. CVPR’08)

Page 23: Edge detection