DOC PREVIEW
CUNY CSC I6716 - Image Segmentation

This preview shows page 1-2-3-4-5-6-42-43-44-45-46-47-86-87-88-89-90-91 out of 91 pages.

Save
View full document
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
View full document
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience
Premium Document
Do you want full access? Go Premium and unlock all 91 pages.
Access to all documents
Download any document
Ad free experience

Unformatted text preview:

Feature ExtractionFinding CirclesRegion SegmentationGoal of SegmentationPrimary Goal of SegmentationRegion Segmentation ExampleColor Image SegmentationProblems with SegmentationFormal Definition of RegionsRepresenting RegionsContour RepresentationChain CodeSlide 13Slide 14Partitioning MethodsGrouping MethodsSlide 17Slide 18Generalized ThresholdingThreshold SelectionSlide 21Optimal ThresholdingLocal or Adaptive ThresholdingSlide 24Improved ResultsLocal Histogram ThresholdingSlide 27Region GrowingSlide 29Similarity CriteriaExample: Simple Region GrowingProblemsSlide 33Region MergingExample Distance MeasuresSlide 36Merging HierarchyFisher’s CriterionHistogram ModelSlide 40UniformitySlide 42Region SplittingRegion splitting: Multiple FeaturesOhlander-Price DataOhlander-Price ResultSlide 47Peak Selection CriteriaSlide 49Hybrid TechniqueSplit and MergeSplit and Merge ExampleSlide 53Segmentation by k-means clusteringObjective Functionk-means algorithmk-means ClusteringSlide 58Slide 59Slide 60Slide 61Slide 62Slide 63Slide 64ExampleResults, k=11Other Distance MeasuresHybrid Edge-Region ApproachesSlide 69Watershed AlgorithmsWatershedsExample on GrayScale ImageImmersion AlgorithmWatershed ProblemsOver-segmentation ProblemTobogganingMultispectral SegmentationColor ImageSegmentationsThree Color UnionFinal Segmentation: After MergingHouse ResultsRegion FeaturesSlide 84RectangularityPerimeterMomentsCentral MomentsOrientationSummaryNext TopicVision, Video and Virtual RealityFeature ExtractionFeature ExtractionLecture 9Image SegmentationCSC 59866CDFall 2004Zhigang Zhu, NAC 8/203Ahttp://www-cs.engr.ccny.cuny.edu/~zhu/Capston e2004/Capst one_Sequence2004.htmlVision, Video and Virtual RealityFinding CirclesFinding CirclesIf we don’t know r, accumulator array is 3-dimensionalIf edge directions are known, computational complexity if reducedSuppose there is a known error limit on the edge direction (say +/- 10o) - how does this affect the search?Hough can be extended in many ways….see, for example:Ballard, D. H. Generalizing the Hough Transform to Detect Arbitrary Shapes, Pattern Recognition 13:111-122, 1981.Illingworth, J. and J. Kittler, Survey of the Hough Transform, Computer Vision, Graphics, and Image Processing, 44(1):87-116, 1988Vision, Video and Virtual RealityRegion SegmentationRegion SegmentationPartitioning of an image into different regions (connected components), each having uniform properties in some (set of) image feature(s):gray valuecolor value(s)textural qualitieslocal gradientmotionshape info..... etc.Vision, Video and Virtual RealityGoal of SegmentationGoal of SegmentationSegment a scene into image elements which may correspond to meaningful scene elementsHigh Level InterpretationsObjectsScene ElementsImage ElementsRaw ImagesVision, Video and Virtual RealityPrimary Goal of SegmentationPrimary Goal of Segmentation“Segmenting an image into image elements which may correspond to meaningful scene elements”What sort of image elements may correspond to meaningful scene elements?Answer depends on type and complexity of images: Less constrained scenes must be segmented more conservatively.Segmentation is not a well defined problem.Vision, Video and Virtual RealityRegion Segmentation ExampleRegion Segmentation ExampleVision, Video and Virtual RealityColor Image SegmentationColor Image SegmentationGiven a grayscale image, how do we generate a region segmentation?In general, regions can be formed from the original image data or from 'derived' images:- color images from R, G, B- textural images- displacement images from motion analyses- 3D depth images?Vision, Video and Virtual RealityProblems with SegmentationProblems with SegmentationIn general, high level contextual knowledge is required for successful segmentationVision, Video and Virtual RealityFormal Definition of RegionsFormal Definition of RegionsA region segmentation of an image, I, is a partition of the set of pixels of I into a set of K regions {Rj}, 1≤j≤K, such that:1. I =”i=1K Rj2. Ri Rj = for i ≠ j3. p connected to p’ for all p, p’ in Rj4. For some predicate PP(Ri) is TRUE for I = 1,2,…,KP(Ri Rj) is FALSE for Ri, Rj adjacent and i≠jEvery pixel belongs to a regionEvery pixel belongs to a regionNo pixel belongs to more than one regionNo pixel belongs to more than one regionSpatial coherenceSpatial coherenceFeature coherenceFeature coherenceVision, Video and Virtual RealityRepresenting RegionsRepresenting RegionsRegion Occupancy MapA set of region labels in registration with image I specifying the region association for each pixel1 1 1 1 11 1 1 1 11 1 11 11 1 112 2 2 2 22 2 2 2 22 2 2 2 22 2 2 2 22 2 2 222 223 3 33 3 34444444444445 5 55 5 555 5 5555 5 5555 5 5555 5 555556666666666666667 7 7 7 7 7777 7 7 77 7 7 7 7 718 8 8 8 88 8 8 8 88 8Image Occupancy Map or Label PlaneVision, Video and Virtual RealityContour RepresentationContour RepresentationC 12C3C4C5C6C7C8C9C10C11C12C13C14C15C16C17C18CR 12R3R4R5R6R7R8R20C19C2C3C4C5C6C7C8C9C10C11C12C13C14C15C16C17C18C2R3R4R5R6R7R8R20C19C1C1RR1 : {C1,C8,C11}R8 : {C1,C5,C17}. . .ImageVision, Video and Virtual RealityChain CodeChain CodeThe chain code representation of a boundary is found by 'walking' counterclockwise around the boundary and recording the direction to turn to stay on the border:3 2 1 4 0 5 6 7Direction CodeVision, Video and Virtual RealityChain CodeChain Code 3 2 1 4 0 5 6 7CC = (i,j) {5 5 6 6 6 0 0 0 0 0 0 0 1 1 2 2 2 4 4 5 4 3 4 4}Vision, Video and Virtual RealityRegion SegmentationRegion SegmentationBasic ApproachesGeneralized thresholdingRegion growingRegion mergingRegion splittingSplit and MergeExtensions to split and mergeK-means clusteringWatershed algorithmsPartitioning methodsGrouping methodsVision, Video and Virtual RealityPartitioning MethodsPartitioning MethodsPartitioning: Given: a large data set. Goal: carve it up according to some notion of the association between items inside the set. We would like to decompose it into pieces that are “good” according to our model. For example, we might:decompose an image into regions which have coherent color and/or texture inside them;take a video sequence and decompose it into shots — segments of video showing about the same stuff from about the same view point;decompose a video sequence into motion


View Full Document

CUNY CSC I6716 - Image Segmentation

Download Image Segmentation
Our administrator received your request to download this document. We will send you the file to your email shortly.
Loading Unlocking...
Login

Join to view Image Segmentation and access 3M+ class-specific study document.

or
We will never post anything without your permission.
Don't have an account?
Sign Up

Join to view Image Segmentation 2 2 and access 3M+ class-specific study document.

or

By creating an account you agree to our Privacy Policy and Terms Of Use

Already a member?