DualBox: Generating BBox Pair with Strong Correspondence via Occlusion Pattern Clustering and Proposal Refinement

Web text exists non-certain and non-structure contents ,and it is difficult to cluster the text by normal classification methods. We propose a web text clustering algorithm based on fuzzy set to increase the computing accuracy with the web text. After abstracting the key words of the text, we can look it as attributes and design the fuzzy algorithm to decide the membership of the words. The algorithm can improve the algorithm complexity of time and space, increase the robustness comparing to the normal algorithm. To test the accuracy and efficiency of the algorithm, we take the comparative experiment between pattern clustering and our algorithm. The experiment shows that our method has a better result.

Download Full-text

Prosodic phrase segmentation based on pitch-pattern clustering

Electronics and Communications in Japan (Part III Fundamental Electronic Science) ◽

10.1002/ecjc.4430770608 ◽

1994 ◽

Vol 77 (6) ◽

pp. 80-91 ◽

Cited By ~ 2

Author(s):

Mitsuru Nakai ◽

Hiroshi Shimodaira ◽

Shigeki Sagayama

Keyword(s):

Pattern Clustering ◽

Pitch Pattern ◽

Prosodic Phrase

Download Full-text

The Design of a Universal Infrared Remote Repeater

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.466-467.1100 ◽

2012 ◽

Vol 466-467 ◽

pp. 1100-1103

Author(s):

Hong Shan Nie ◽

Qiang Liu ◽

Miao Li ◽

Qing Jiang Li ◽

Hai Jun Liu ◽

...

Keyword(s):

Pattern Recognition ◽

Remote Control ◽

Vector Quantization ◽

Control Signal ◽

Good Prospect ◽

Clock Signal ◽

Storage Efficiency ◽

Pattern Clustering ◽

Decoding Efficiency ◽

Repetitive Pattern

In this paper, a common infrared remote control transmitter is designed, using which, a variety of infrared remote control signal can be decoded, stored and transmitted when users need. And with this design, the problem that remote controllers can not be universally operated is solved . In the design, the algorithm of pattern clustering is used to reduce the errors caused by the instability of clock signal. Moreover, the method of vector Quantization coding and repetitive pattern recognition is used to compress he remote control signal data which achieve a good decoding efficiency and a storage efficiency. Experiments show that the design is feasible and has a good prospect in application.

Download Full-text

WIS-ART: UNSUPERVISED CLUSTERING WITH RAM DISCRIMINATORS

International Journal of Neural Systems ◽

10.1142/s0129065792000061 ◽

1992 ◽

Vol 03 (01) ◽

pp. 57-63 ◽

Cited By ~ 3

Author(s):

Eamon P. Fulcher

Keyword(s):

Random Access ◽

Test Phase ◽

Adaptive Resonance Theory ◽

The Self ◽

Hierarchical Level ◽

Resonance Theory ◽

Adaptive Resonance ◽

Pattern Clustering ◽

Adaptive Pattern ◽

System Operating

WIS-ART merges the self-organising properties of Adaptive Resonance Theory (ART) with the operation of WISARD, an adaptive pattern recognition machine which uses discriminators of conventional Random Access Memories (RAMs). The result is an unsupervised pattern clustering system operating at near real-time that implements the leader algorithm. ART’s clustering is highly dependent upon the value of a “vigilance” parameter, which is set prior to training. However, for WIS-ART hierarchical clustering is performed automatically by the partitioning of discriminators into “multi-vigilance modules”. Thus, clustering may be controlled during the test phase according to the degree of discrimination (hierarchical level) required. Methods for improving the clustering characteristics of WIS-ART whilst still retaining stability are discussed.

Download Full-text