30.01.2013 Views

Shih_Image_Processing_and_Mathematical_Morpholo.pdf

Shih_Image_Processing_and_Mathematical_Morpholo.pdf

Shih_Image_Processing_and_Mathematical_Morpholo.pdf

SHOW MORE
SHOW LESS

Create successful ePaper yourself

Turn your PDF publications into a flip-book with our unique Google optimized e-Paper software.

xxii Author<br />

Dr. <strong>Shih</strong> has contributed as a steering member, committee member, <strong>and</strong><br />

session chair for numerous professional conferences <strong>and</strong> workshops. He is<br />

presently on the Technical Committee of the IEEE Circuits <strong>and</strong> Systems<br />

Society in Visual Signal <strong>Processing</strong> & Communications <strong>and</strong> Multimedia<br />

Systems & Applications. He was the recipient of the Research Initiation<br />

Award from the National Science Foundation in 1991. He won the Honorable<br />

Mention Award from the International Pattern Recognition Society for an<br />

outst<strong>and</strong>ing paper <strong>and</strong> also won the Best Paper Award at the International<br />

Symposium on Multimedia Information <strong>Processing</strong>. Dr. <strong>Shih</strong> has received<br />

numerous grants from the National Science Foundation, the navy <strong>and</strong> air<br />

force, <strong>and</strong> industry. He has received several research merit awards at the<br />

New Jersey Institute of Technology. He has served several times on the<br />

Proposal Review Panel of the National Science Foundation.<br />

Professor <strong>Shih</strong> was one of the researchers who initiated mathematical morphology<br />

research with applications to image processing, feature extraction,<br />

<strong>and</strong> object representation. His IEEE PAMI article “Threshold decomposition of<br />

grayscale morphology into binary morphology” is a breakthrough paper that<br />

solves the bottleneck problem in grayscale morphological processing. His<br />

several IEEE <strong>Image</strong> <strong>Processing</strong> <strong>and</strong> IEEE Signal <strong>Processing</strong> articles detail innovations<br />

that achieve fast exact Euclidean distance transformation, robust image<br />

enhancement, edge linking, <strong>and</strong> image segmentation using the recursive soft<br />

morphological operators, general sweep morphological operators, alternating<br />

sequential morphological fi lters, <strong>and</strong> regulated morphological fi lters that<br />

he developed.<br />

In the fi eld of digital document processing, Professor <strong>Shih</strong> developed a<br />

fuzzy model for unsupervised character classifi cation, fuzzy typographical<br />

analysis for character classifi cation, an adaptive system for block segmentation<br />

<strong>and</strong> classifi cation, <strong>and</strong> a rule-based character <strong>and</strong> logo recognition system.<br />

In the fi eld of face recognition, Professor <strong>Shih</strong> developed automatic<br />

extraction of head <strong>and</strong> face boundaries <strong>and</strong> facial features, a hybrid twophase<br />

algorithm for face recognition, multiview face identifi cation <strong>and</strong> pose<br />

estimation using B-spline interpolation, recognizing facial action units using<br />

independent component analysis <strong>and</strong> a support vector machine, <strong>and</strong> facial<br />

expression recognition.<br />

In the fi eld of pattern recognition, Professor <strong>Shih</strong> developed improved feature<br />

reduction in input <strong>and</strong> feature spaces, an improved incremental training<br />

algorithm for support vector machines using active query, support vector<br />

machine networks for multiclass classifi cation, improved adaptive resonance<br />

theory networks, model-based partial shape recognition using contour<br />

curvature <strong>and</strong> affi ne transformation, <strong>and</strong> a distance-based separator<br />

representation for pattern classifi cation. In the fi eld of image segmentation,<br />

Professor <strong>Shih</strong> developed automatic seeded region growing for color image<br />

segmentation, locating object contours in complex backgrounds using an<br />

improved snake model, a top-down region-dividing approach for image segmentation,<br />

<strong>and</strong> edge linking by adaptive mathematical morphology.

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!