Shape and Texture Recognition by a Neural Network

Khotanzad, A.; Lu, J.-H.

doi:10.1016/b978-0-444-88740-5.50012-8

Cited by 12 publications

(5 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(Using nearest-neighbors as a benchmark is common see e.g., Geman et al (1992), Khotanzad & Lu (1991)). Let \NN(raw)" refer to nearest-neighbor classi cation based on Hamming distance in (binary) image space, i.e., between bitmaps.…”

Section: Generalizationmentioning

confidence: 99%

“…Many i n variants for planar objects (based on single views) and for non-planar objects (based on multiple views) have been discovered and proposed for recognition see Reiss (1993) and the references therein. Some invariants are based on Fourier descriptors and image moments for example, the magnitude of Zernike m o m e n ts (Khotanzad & Lu (1991)) are invariant t o r otation. Most invariants require computing tangents from estimates of the shape boundaries (Forsyth, Mundy, Zisserman, Coelho, Heller & Rothwell (1991), Sabourin & Mitiche (1992)).…”

Section: Invariant Recognitionmentioning

confidence: 99%

See 1 more Smart Citation

Shape Quantization and Recognition with Randomized Trees

1997

View full text Add to dashboard Cite

We explore a new approach to shape recognition based on a virtually in nite family of binary features (\queries") of the image data, designed to accommodate prior information about shape invariance and regularity. E a c h query corresponds to a spatial arrangement o f s e v eral local topographic codes (\tags") which are in themselves too primitive and common to be informative about shape. All the discriminating power derives from relative angles and distances among the tags. The important attributes of the queries are (i) a natural partial ordering corresponding to increasing structure and complexity (ii) semi-invariance, meaning that most shapes of a given class will answer the same way t o t wo queries which are successive in the ordering and (iii) stability, since the queries are not based on distinguished points and substructures.No classi er based on the full feature set can be evaluated and it is impossible to determine a priori which arrangements are informative. Our approach i s t o s e l e c t informative features and build tree classi ers at the same time by inductive learning. In e ect, each tree provides an approximation to the full posterior where the features chosen depend on the branch which i s t r a versed. Due to the number and nature of the queries, standard decision tree construction based on a xed length feature vector is not feasible. Instead we e n tertain only a small random sample of queries at each n o d e , constrain their complexity to increase with tree depth, and grow m ultiple trees. The terminal nodes are labeled by estimates of the corresponding posterior distribution over shape classes. An image is classi ed by sending it down every tree and aggregating the resulting distributions.The method is applied to classifying handwritten digits and synthetic linear and nonlinear deformations of three hundred Latex symbols. State-of-the-art error rates are achieved on the NIST database of digits. The principal goal of the experiments on Latex symbols is to analyze invariance, generalization error and related issues, and a comparison with ANN methods is presented in this context. 2

show abstract

Section: Generalizationmentioning

confidence: 99%

Section: Invariant Recognitionmentioning

confidence: 99%

Shape Quantization and Recognition with Randomized Trees

1997

View full text Add to dashboard Cite

show abstract

“…In Khotanzad and Lu [15] the application of MLP classifiers to texture recognition is considered, and its superiority over other traditional classifiers is demonstrated.…”

Section: Other Applicationsmentioning

confidence: 99%

Application of multi-layer perceptron neural networks to vision problems

Khotanzad

Chung

1998

Neural Comput & Applic

View full text Add to dashboard Cite

show abstract

“…These applications include speech recognition (Bengio, 1993), human face recognition (Soulie et al, 1993), satellite image classification (Civco, 1993;Atkinson and Tatnall, 1997;Bandibas and Kohyama, 2001) and shape and texture recognition (Khotanzad and Lu, 1991). One of the advantages of using an ANN for qualitative modeling of natural phenomena is that it can handle data at any measurement scale ranging from nominal, ordinal to linear and ratio, and any form of data distribution (Wang et al, 1995).…”

Section: Introductionmentioning

confidence: 99%