49
Image Analysis: Object Recognition

Image Analysis: Object Recognition

Embed Size (px)

DESCRIPTION

Image Analysis: Object Recognition. Image Segmentation. Image Analysis: Object Recognition. INPUT IMAGE. OBJECT IMAGE. Image Segmentation: each object in the image is identified and isolated from the rest of the image. Feature Extraction. Image Analysis: Object Recognition. x - PowerPoint PPT Presentation

Citation preview

Page 1: Image Analysis:  Object Recognition

Image Analysis: Object Recognition

Page 2: Image Analysis:  Object Recognition

Image Segmentation

INPUT IMAGE

OBJECT IMAGE

Image Segmentation: each object in the image is identified and isolated from the rest of the image

Image Analysis: Object Recognition

Page 3: Image Analysis:  Object Recognition

Feature Extraction

OBJECT IMAGE

FEATURE VECTORS

xxx…x

Feature Extraction: measurements or “features” are computed on each object identified during the segmentation step

Image Analysis: Object Recognition

1

2

3

n

Page 4: Image Analysis:  Object Recognition

x 1

x 2

x n

The feature vector for a given pixel consists of the corresponding pixels from each feature image; the feature vector for an object would be computed from pixels comprising the object, from each feature image.

Page 5: Image Analysis:  Object Recognition

Classification: each object is assigned to a class

FEATURE VECTORS

Image Analysis: Object Recognition

Classification

OBJECT TYPE

“WRENCH”

Page 6: Image Analysis:  Object Recognition

Image Analysis: Object Recognition

Image Segmentation

Feature Extraction

Classification

INPUT IMAGE OBJECT IMAGE

FEATURE VECTOR

OBJECT TYPE

“WRENCH”

Page 7: Image Analysis:  Object Recognition

Example: an automated fruit sorting system

Page 8: Image Analysis:  Object Recognition

Example: an automated fruit sorting system

segmentation: identify the fruit objects

the image is partitioned to isolate individual fruit objects

Page 9: Image Analysis:  Object Recognition

Example: an automated fruit sorting system

segmentation: identify the fruit objects

feature extraction: compute a size and color feature for each segmented region in the image

size - diameter of each objectcolor - red-to-green brightness ratio

(redness measure)

Page 10: Image Analysis:  Object Recognition

Example: an automated fruit sorting system

segmentation: identify the fruit objects

feature extraction: compute a size and color feature for each segmented region in the image

classification: partition the “fruit” objects in feature space

Page 11: Image Analysis:  Object Recognition
Page 12: Image Analysis:  Object Recognition

Automatic (unsupervised) image Segementation : difficult problem

1) attempt to control imaging conditions (industrial applications)

2) choose sensor which enhance objects of interest(infared imaging)

Page 13: Image Analysis:  Object Recognition

Segmentation Algorithms:

- discontinuities between homogeneous regions

- similarity of pixel values within a region

Page 14: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

detect points, lines and edges in an image

Page 15: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

detect points, lines and edges in an image

-1 -1 -1-1 8 -1-1 -1 -1

Page 16: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

detect points, lines and edges in an image

-1 -1 -1-1 8 -1-1 -1 -1

-1 -1 -1 2 2 2-1 -1 -1

-1 2 -1-1 2 -1-1 2 -1

-1 -1 2-1 2 -1 2 -1 -1

2 -1 -1-1 2 -1-1 -1 2

Page 17: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

detect points, lines and edges in an image

-1 -1 -1-1 8 -1-1 -1 -1

-1 -1 -1 2 2 2-1 -1 -1

-1 2 -1-1 2 -1-1 2 -1

-1 -1 2-1 2 -1 2 -1 -1

2 -1 -1-1 2 -1-1 -1 2

-1 -2 -1 0 0 0 1 2 1

-1 0 1-2 0 2-1 0 1

Page 18: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

detect points, lines and edges in an image

-1 -2 -1 0 0 0 1 2 1

-1 0 1-2 0 2-1 0 1

Gx Gy

Page 19: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

detect points, lines and edges in an image

-1 -2 -1 0 0 0 1 2 1

-1 0 1-2 0 2-1 0 1

Gx Gy

Page 20: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

GxGy

Gradient vector

Edge Linking - used to create connected boundaries

Page 21: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

GxGy

Gradient vector

Edge Linking - used to create connected boundaries

- similar points within a neighborhood are linked

Page 22: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

GxGy

Gradient vector

Edge Linking - used to create connected boundaries

- similar points within a neighborhood are linked

magnitude of gradient vector

[ Gx + Gy ]2 22

1

Page 23: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

GxGy

Gradient vector

Edge Linking - used to create connected boundaries

- similar points within a neighborhood are linked

magnitude of gradient vector

[ Gx + Gy ]

approximated as | Gx | + | Gy |

2 22

1

Page 24: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

GxGy

Gradient vector

Edge Linking - used to create connected boundaries

- similar points within a neighborhood are linked

magnitude of gradient vector

orientation of edges

ang(x,y) = tan ( )-1 Gy

Gx

Page 25: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

GxGy

Gradient vector

Edge Linking - used to create connected boundaries

- similar points within a neighborhood are linked

magnitude of gradient vector

orientation of edges

Page 26: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

Identify zero crossings

Page 27: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

Identify zero crossings

0 -1 0-1 4 -1 0 -1 0

Page 28: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

Identify zero crossings

Page 29: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

Identify zero crossings

Page 30: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

Identify zero crossings

Page 31: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

Identify zero crossings

Page 32: Image Analysis:  Object Recognition

Discontinuity based Segmentation:

Identify zero crossings

Page 33: Image Analysis:  Object Recognition

Similarity based Segmentation:

- Simple thresholding- Split and Merge- Recursive thresholding

Page 34: Image Analysis:  Object Recognition

Similarity based Segmentation:

- Simple thresholding- Split and Merge- Recursive thresholding

Page 35: Image Analysis:  Object Recognition

Single Level Thresholding

0, g < TH

G - 1, TH gT[g] =

Page 36: Image Analysis:  Object Recognition

Single Level Thresholding

0, g < TH

G - 1, TH gT[g] =

Page 37: Image Analysis:  Object Recognition

Single Level Thresholding

Page 38: Image Analysis:  Object Recognition

Single Level Thresholding

0, g < TH

G - 1, TH gT[g] =

Page 39: Image Analysis:  Object Recognition

Multiple Level Thresholding

0, g < TH1

G - 1, TH1 g <= TH2

0, g > TH2

T[g] =

Page 40: Image Analysis:  Object Recognition

Similarity based Segmentation:

- Simple thresholding- Split and Merge- Recursive thresholding

Page 41: Image Analysis:  Object Recognition

U

Split and Merge

1) split region into four disjoint quadrants if P(Rj) = FALSE

2) merge any adjacent regions Rj and Rk if P(Rj Rk) = TRUE

3) stop when no splitting or merging is possible

Page 42: Image Analysis:  Object Recognition

Split and Merge

Page 43: Image Analysis:  Object Recognition

Split and Merge

Page 44: Image Analysis:  Object Recognition

Split and Merge

Page 45: Image Analysis:  Object Recognition

Split and Merge

Page 46: Image Analysis:  Object Recognition

Split and Merge

Page 47: Image Analysis:  Object Recognition

Split and Merge

Page 48: Image Analysis:  Object Recognition

Split and Merge

Page 49: Image Analysis:  Object Recognition

Split and Merge