Color measuring system using multispectral image data

• US 6,269,182 B1
• Filed: 05/16/1996
• Issued: 07/31/2001
• Est. Priority Date: 05/23/1995
• Status: Expired due to Fees

• Current Assignee: Olympus Optical Company Ltd. (Olympus Corporation)
• Original Assignee: Olympus Optical Company Ltd. (Olympus Corporation)
• Inventors: Ishii, Kensuke
• Primary Examiner(s): CHANG, JON CARLTON
• Application Number:
  US08/649,037
• Time in Patent Office: 1,902 Days
• Field of Search: 382/165, 382/162, 382/224, 382/225, 382/226, 382/228, 382/227, 364/526, 356/402, 356/406, 356/418, 356/419
• US Class Current: 382/165
• CPC Class Codes: G01J 2003/466 Coded colour; Recognition of predetermined colour; Determining proximity to predetermined colour
• CPC Class Codes: G01J 2003/467 Colour computing
• CPC Class Codes: G01J 3/462 Computing operations in or between colour spaces; Colour management systems
• CPC Class Codes: G01J 3/51 using colour filters

Abstract

A color measuring system includes a color classifying apparatus and a color nonuniformity checking apparatus capable of measuring color nonuniformity which simplify the configuration, reduce cost, withstand mechanical vibration, and are capable of classifying the colors of objects well without limiting the light source even when its spectrum varies, and which are capaple of further improving the color classification accuracy. The color classifying apparatus includes an imaging section for imaging the reflected light from an object, a plurality of band-pass dilters each having a different band and provided between the object and the imaging section, and a classifying section for computing a classification spectrum for classification using a statistical approach from the multispectral image data on the object imaged by the imaging section and classifying the object using the classification spectrum, wherein the classifying section uses a classification judgment method most suitable for classification judgement on the object.

Claims

1. A color classifying apparatus comprising:

imaging means for imaging reflected light from an object;

optical means for forcing said reflected light from the object to form multispectral images each having a different band on said imaging means; and

classifying means for computing a classification spectrum for classification using a statistical approach based on multispectral image data relating to the multispectral images formed by said imaging means, and for classifying said object into a plurality of classes using the computed classification spectrum,

wherein said classifying means includes;

(i) a plurality of classification judging sections that make a classification judgment on said plurality of classes using a plurality of different classification judgment methods; and

(ii) a judgment result judging section that comprehensively judges the classification judgments of one or more of said plurality of classification judging sections, and

wherein said judgment result judging section of said classifying means judges the classification judgment of a first one of said plurality of classification judging sections and then judges whether to cause a second one of said plurality of classification judging sections to make a classification judgment based on a judgment result of the classification judgment of the first one of said classification judging sections.

2. A color classifying apparatus according to claim 1, wherein said plurality of classification judgment sections are connected in series and make classification judgments in a superimposing manner.

3. A color classifying apparatus according to claim 1, wherein said classifying means further includes:

a class information database in which class information on said object has been stored; and

a classification judgment selecting function section that selects a classifying process and a narrowing-down method for said plurality of classification judging sections according to the class information stored in said class information database.

4. A color classifying apparatus according to claim 1, wherein said classifying means further includes a learning control section that generates learning data based on said multispectral image data, and that controls a classification process of said classifying means according to the generated learning data.

5. A color classifying apparatus according to claim 1, wherein said classifying means further includes a learning data updating section that updates learning data when said classification judgments are being made.

6. A color classifying apparatus comprising:

imaging means for imaging reflected light from an object;

optical means for forcing said reflected light from the object to form multispectral images each having a different band on said imaging means; and

classifying means for computing a classification spectrum for classification using a statistical approach based on multispectral image data relating to the multispectral images formed by said imaging means, and for classifying said object into a plurality of classes using the computed classification spectrum,

wherein said classifying means includes;

(i) a plurality of classification judging sections that make a classification judgment on said plurality of classes using a plurality of different classification judgment methods; and

(ii) a judgment result judging section that comprehensively judges the classification judgments of one or more of said plurality of classification judging sections, and

wherein first and second ones of said plurality of classification judging sections make initial classification judgments in a superimposing manner, and said judgment result judging section then judges whether to cause said second one of said plurality of classification judging sections to make another classification judgment according to a judgment result of the initial classification judgment of said second one of said plurality of classification judging sections.

7. A color classifying apparatus according to claim 6, wherein said plurality of classification judgment sections are connected in series and make classification judgments in a superimposing manner.

8. A color classifying apparatus according to claim 6, wherein said classifying means further includes:

a class information database in which class information on said object has been stored; and

a classification judgment selecting function section that selects a classifying process and a narrowing-down method for said plurality of classification judging sections according to the class information stored in said class information database.

9. A color classifying apparatus according to claim 6, wherein said classifying means further includes a learning control section that generates learning data based on said multispectral image data, and that controls a classification process of said classifying means according to the generated learning data.

10. A color classifying apparatus according to claim 6, wherein said classifying means further includes a learning data updating section that updates learning data when said classification judgments are being made.

11. A color classifying apparatus comprising:

imaging means for imaging reflected light from an object;

optical means for forcing said reflected light from the object to form multispectral images each having a different band width on said imaging means;

memory means for storing a plurality of sets of classification spectrums for classifying said reflected light from the object using a statistical approach based on multispectral image data relating to the multispectral images formed by said imaging means; and

classifying means for classifying said reflected light from the object based on said plurality of is sets of classification spectrums stored in said memory means;

wherein said classifying means includes;

(i) first classification judging means for firstly judging said multispectral image data, and classifying the multispectral image data for narrowing down to a plurality of vicinal classes; and

(ii) second classification judging means for rejudging the multispectral image data in each of said plurality of vicinal classes, which are narrowed down by said first classification means, within each of the vicinal classes, and for outputting a classification result.

12. A color classifying apparatus according to claim 11, wherein said classifying means further includes a learning control section that generates learning data based on said multispectral image data, and that controls a classification process of said classifying means according to the generated learning data.

13. A color classifying apparatus according to claim 11, wherein said classifying means further includes a learning data updating section that updates learning data when a classification judgment is being made.