Unlock instant, AI-driven research and patent intelligence for your innovation.

A classification system

A classification system, classifier technology, applied in the direction of instruments, character and pattern recognition, computer parts, etc., can solve problems such as increasing training time

Inactive Publication Date: 2006-11-08
MV RES
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Unfortunately, unilateral classifiers are only as accurate as bilateral classifiers under very specific and few conditions
Furthermore, it is empirically found that unilateral classifiers may require more concept training data than bilateral classifiers, thereby significantly increasing training time

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A classification system
  • A classification system
  • A classification system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0022] The present invention provides an effective classifier for machine vision systems when only concept samples for specific inspection tasks are on hand. It uses a library of counter-concept samples from some similar tasks.

[0023] Generally, counter-concept samples are likely to be discovered by the inspection system during its operation. Since many inspection tasks require repeated creation of classifiers for similar types of problems, it is reasonable to infer that in these situations, a considerable sample library of similar anti-concepts can be collected. For example, for solder joint inspection, it may be necessary to establish a separate classifier for each joint type on the board of each model manufactured. All contacts marked as defective by these respective systems can be collected in a sample library of defective contacts.

[0024] reference figure 1 The filter process 1 allows a subset of the previously collected counter-concept sample library 2 to be selected, a...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A classification system or classifying images of solder joints has a two-sided classifier to which query cases are presented. It can operate in an environment without task-specific counter-concept cases ('bad examples') because it filters a library of general counter-concept cases to provide a refined set of counter-concept cases. The filtering is performed by a one-sided classifier, which uses a base of task-specific concept cases to perform the filtration.

Description

Technical field [0001] The present invention relates to a classification system and a machine vision system including the classification system. Background technique [0002] In general, binary classifiers are trained to distinguish between concept classes and counter-concept classes. Training includes providing a set of samples that are known to be correctly classified to the classifier, and adjusting internal parameters based on the ability of the classifier to correctly classify these training samples. When both concept and counter-concept samples are used during training, the classifier is called two-sided, and the classification accuracy is generally high. [0003] However, in some cases, anti-concept training samples cannot be obtained or it is expensive to obtain anti-concept samples, so the classifier must be trained without anti-concept samples. An example of this is a classifier used as part of a machine vision system for solder joint inspection (ie distinguishing betwe...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/62
CPCG06V2201/06G06F18/243G06F18/214
Inventor 詹姆士·马洪布雷恩·麦克内密约翰·多尔蒂理查德·艾万斯
Owner MV RES