36
A Wrapper-Based Approach to Image Segmentation and Classification Michael E. Farmer, Member, I EEE, and Anil K. Jain, Fellow, IEEE

A Wrapper-Based Approach to Image Segmentation and Classification Michael E. Farmer, Member, IEEE, and Anil K. Jain, Fellow, IEEE

  • View
    213

  • Download
    0

Embed Size (px)

Citation preview

A Wrapper-Based Approach to Image Segmentation and Classification

Michael E. Farmer, Member, IEEE, and Anil K. Jain, Fellow, IEEE

大綱大綱 IntroductionIntroduction Overview of the approachOverview of the approach Experiment: Vision-Base airbag suppression Experiment: Vision-Base airbag suppression

applicationapplication Experimental resultExperimental result

IntroductionIntroduction

Traditional processingTraditional processing

The traditional processing flow for image-based pattern recognition consists of image segmentation followed by classification.

Three limitations of traditional Three limitations of traditional processingprocessing

The object of interest “should be uniform and homogeneous with respect to some characteristic” and “adjacent regions should be differing significantly”

There are few metrics available for evaluating segmentation algorithms

Inability to adapt to real-world changes

The contributions in this paper

Developing a closed-loop framework for image segmentation to find the best segmentation for a given class of objects by using the shape of the object for classification of the segmented object

Using the probability of correct classification of the object to provide an “objective evaluation of segmented outputs”

The system can adapt to “real-world changes.”

Overview of the approachOverview of the approach

Wrapper-Based ApproachWrapper-Based Approach

Wrap the segmentation and the classification together, and use the classifier as the metric for selecting the best segmentation.

Using the classifier to intelligently re-assemble to solve over-segmented problem.

The classification is correct when the minimum distance between the classification of the candidate segmentation and one of the desired pattern classes < T

Traditional vs Wrapper-BaseTraditional vs Wrapper-Base

Experiment: Experiment: Vision-Base airbag suppression Vision-Base airbag suppression

applicationapplication

ProblemProblem

Infant or Adult

ChallengesChallenges

Nonuniform illumination Poor image contrast Shadows and highlights Occlusions Sensor noise Background clutter

Variability for the infant classVariability for the infant class

Variability for the infant classVariability for the infant class

Proposed approachProposed approach

Preliminary SegmentationPreliminary Segmentation Reduce the number of blobs that must be processed.

Once the correlation value for each region is determined, an adaptive threshold is applied, and any region that falls below the threshold is considered a part of the foreground.

Preliminary SegmentationPreliminary Segmentation

Preliminary SegmentationPreliminary Segmentation

RegionRegion LabelingLabeling

Using the EM algorithm with a fixed number of components, and then rely on the classification accuracy to determine if more components are required.

Merging the very small blobs by mode filter Merging any regions that are smaller then 20

pixels in size with their larger neighbors

RegionRegion Labeling ResultsLabeling Results

RegionRegion Labeling ResultsLabeling Results

Blob CombinerBlob Combiner

We have framed the blob combiner problem as one of blob selection, where there exists a subset of blobs that will provide the highest classification accuracy for a given pattern class

Forward selection modeForward selection mode Backward selection modeBackward selection mode

Blob CombinerBlob Combiner( ( plus-L, minus-R algorithm )

Blob CombinerBlob Combiner ( ( plus-L, minus-R algorithm )

Feature ExtractionFeature Extraction

Feature ExtractionFeature Extraction

Acceleration Methods forAcceleration Methods for Feature Extraction Feature Extraction:

Precompute the moments for each blob Compute the moments using only the local nei

ghborhood of each blob.

Attain over a ten thousand-fold reduction in processing for each moment calculated.

Classification of Blob CombinationsClassification of Blob Combinations

Using the nearest neighbor classifier to compute classification distance

Feature 1

Fea

ture

2

: class - A points: class - B points: point with unknown class

Circle of 1 - nearest neighborThe point is class B via 1-NNR.

Proposed approachProposed approach

Demonstrating

Demonstrating

Demonstrating

EXPERIMENTAL RESULTS

EXPERIMENTAL RESULTS

Correct segmentations

Incorrect segmentation