Patent · US11067501B2 · B2 · US

Fabric validation using spectral measurement

(11) Publication number

US11067501B2

(21) Application number

16/596,997

(22) Filing date

2019-10-09

(30) Priority date

2019-03-29

(43) Publication date

2021-07-20

(45) Date of grant

2021-07-20

(51) IPC

G01J 3/02; G01N 21/3563; G01N 21/359; G01J 3/28; G01J 3/46; G01N 33/36; G06K 9/00; G06N 3/08

(52) CPC

G01N Investigating or analysing materials by determining their chemical or physical properties: 21/3563, 21/359, 2201/0221, 2201/129, 2201/1296, 33/367
G01J Measurement of intensity, velocity, spectral content, polarisation, phase or pulse characteristics of infrared, visible or ultraviolet light; colorimetry; radiation pyrometry: 3/0264, 3/0272, 3/0291, 3/0297, 3/28, 3/42, 3/462
G06F Electric digital data processing: 2218/12
G06K Graphical data reading; presentation of data; record carriers; handling record carriers: 9/00536
G06N Computing arrangements based on specific computational models: 3/0464, 3/08, 3/09

(73) Assignee

Inspectorio Inc

(72) Inventors

Binh Thanh Nguyen; Han Ky Cao; Cuong Van Nguyen; Carlos Moncayo

(54) Title

Fabric validation using spectral measurement

(57) Abstract

Fabric validation using spectral measurement is provided. In various embodiments, a near-infrared absorption spectrum of a fabric sample is received from a near-infrared spectrometer. A plurality of features is extracted from the spectrum. The plurality of features is provided to a trained classifier. The trained classifier provides a similarity score indicative of the similarity of the fabric sample to a reference fabric sample.

Full text

View on Google Patents

Claims (36)

A system comprising: a near-infrared spectrometer; a computing node operatively coupled to the near-infrared spectrometer, the computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform a method comprising: receiving a near-infrared absorption spectrum of a fabric sample from the near-infrared spectrometer; extracting a plurality of features from the spectrum; generating at least one fixed-size vector from the plurality of features; providing the at least one fixed-size vector to a trained classifier; obtaining from the trained classifier a similarity score indicative of the similarity of the fabric sample to a reference fabric sample.
The system of claim 1, wherein the near-infrared spectrometer comprises a hand-held spectrometer.
The system of claim 1, wherein the computing node is local to the near-infrared spectrometer.
The system of claim 3, wherein the computing node and the near-infrared spectrometer are operatively coupled via a local area network.
The system of claim 4, wherein the local area network comprises a wireless network.
The system of claim 3, wherein the computing node and the near-infrared spectrometer are operatively coupled via a personal area network.
The system of claim 3, wherein the computing node and the near-infrared spectrometer are integrated into a handheld device.
The system of claim 1, wherein providing the at least one fixed-size vector to the trained classifier comprises sending the at least one fixed-size vector to a remote fabric validation server, and obtaining from the trained classifier the similarity score comprises receiving the similarity score from the fabric validation server.
The system of claim 8, wherein said sending and receiving is performed via a wide area network.
The system of claim 1, wherein extracting a plurality of features from the spectrum comprises noise reduction.
The system of claim 1, wherein the method further comprises: receiving wavelength, intensity, and/or reflectance from the near-infrared spectrometer, and wherein extracting the plurality of features further comprises extracting features from the wavelength, intensity, and/or reflectance.
The system of claim 1, wherein the trained classifier comprises an artificial neural network.
The system of claim 1, wherein extracting the plurality of features comprises principal component analysis.
The system of claim 1, wherein extracting the plurality of features comprises applying an artificial neural network.
The system of claim 14, wherein the artificial neural network comprises at least one convolutional layer.
The system of claim 15, wherein the artificial neural network comprises a plurality of 1D convolutional layers.
The system of claim 1, wherein the method further comprises obtaining from the trained classifier a fabric material composition, weave type, thread count, yarn thickness, and/or color.
The system of claim 1, wherein obtaining the similarity score comprises providing at least one fixed-size vector extracted from a reference sample to the trained classifier.
The system of claim 1, the method further comprising providing the similarity score to a user.
The system of claim 17, further comprising providing the fabric material composition, weave type, thread count, yarn thickness, and/or color to a user.
A method comprising: receiving a near-infrared absorption spectrum of a fabric sample from a near-infrared spectrometer; extracting a plurality of features from the spectrum; generating at least one fixed-size vector from the plurality of features; providing the at least one fixed-size vector to a trained classifier; obtaining from the trained classifier a similarity score indicative of the similarity of the fabric sample to a reference fabric sample.
The method of claim 21, wherein the near-infrared spectrometer comprises a hand-held spectrometer.
The method of claim 21, wherein providing the at least one fixed-size vector to the trained classifier comprises sending the at least one fixed-size vector to a remote fabric validation server, and obtaining from the trained classifier the similarity score comprises receiving the similarity score from the fabric validation server.
The method of claim 23, wherein said sending and receiving is performed via a wide area network.
The method of claim 21, wherein extracting a plurality of features from the spectrum comprises noise reduction.
The method of claim 21, further comprising: receiving wavelength, intensity, and/or reflectance from the near-infrared spectrometer, and wherein extracting the plurality of features further comprises extracting features from the wavelength, intensity, and/or reflectance.
The method of claim 21, wherein the trained classifier comprises an artificial neural network.
The method of claim 21, wherein extracting the plurality of features comprises principal component analysis.
The method of claim 21, wherein extracting the plurality of features comprises applying an artificial neural network.
The method of claim 29, wherein the artificial neural network comprises at least one convolutional layer.
The method of claim 30, wherein the artificial neural network comprises a plurality of 1D convolutional layers.
The method of claim 21, further comprising obtaining from the trained classifier a fabric material composition, weave type, thread count, yarn thickness, and/or color.
The method of claim 21, wherein obtaining the similarity score comprises providing at least one fixed-size vector extracted from a reference sample to the trained classifier.
The method of claim 21, the method further comprising providing the similarity score to a user.
The method of claim 32, further comprising providing the fabric material composition, weave type, thread count, yarn thickness, and/or color to a user.
A computer program product for fabric validation, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising: receiving a near-infrared absorption spectrum of a fabric sample from a near-infrared spectrometer; extracting a plurality of features from the spectrum; generating at least one fixed-size vector from the plurality of features; providing the at least one fixed-size vector to a trained classifier; obtaining from the trained classifier a similarity score indicative of the similarity of the fabric sample to a reference fabric sample.

Description

Embodiments of the present disclosure relate to fabric validation, and more specifically, to fabric validation using spectral measurement.

According to embodiments of the present disclosure, methods of and computer program products for fabric validation are provided. In various embodiments, a near-infrared absorption spectrum of a fabric sample is received from a near-infrared spectrometer. A plurality of features is extracted from the spectrum. The plurality of features is provided to a trained classifier. The trained classifier provides a similarity score indicative of the similarity of the fabric sample to a reference fabric sample.

In various embodiments, the near-infrared spectrometer comprises a hand-held spectrometer. In various embodiments, providing the plurality of features to the trained classifier comprises sending the plurality of features to a remote fabric validation server, and obtaining from the trained classifier the similarity score comprises receiving the similarity score from the fabric validation server. In various embodiments, said sending and receiving is performed via a wide area network.

In various embodiments, extracting a plurality of features from the spectrum comprises noise reduction.

In various embodiments, wavelength, intensity, and/or reflectance are received from the near-infrared spectrometer, and extracting the plurality of features further comprises extracting features from the wavelength, intensity, and/or reflectance.

In various embodiments, the trained classifier comprises an artificial neural network.

Citations (15)

US6272479B1
US7617163B2
US7071469B2
US20040119972A1
US20100036795A1
US8081304B2
US8452716B2
US20100290032A1
US20170032285A1
US10307795B2
US20200320769A1
US10936921B2
US20200249085A1
WO2020109170A1
US20200042822A1

Record as JSON

{
  "publication_number": "US11067501B2",
  "country": "US",
  "kind": "B2",
  "title": "Fabric validation using spectral measurement",
  "abstract": "Fabric validation using spectral measurement is provided. In various embodiments, a near-infrared absorption spectrum of a fabric sample is received from a near-infrared spectrometer. A plurality of features is extracted from the spectrum. The plurality of features is provided to a trained classifier. The trained classifier provides a similarity score indicative of the similarity of the fabric sample to a reference fabric sample.",
  "claims": [
    "1. A system comprising: a near-infrared spectrometer; a computing node operatively coupled to the near-infrared spectrometer, the computing node comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor of the computing node to cause the processor to perform a method comprising: receiving a near-infrared absorption spectrum of a fabric sample from the near-infrared spectrometer; extracting a plurality of features from the spectrum; generating at least one fixed-size vector from the plurality of features; providing the at least one fixed-size vector to a trained classifier; obtaining from the trained classifier a similarity score indicative of the similarity of the fabric sample to a reference fabric sample.",
    "2. The system of claim 1, wherein the near-infrared spectrometer comprises a hand-held spectrometer.",
    "3. The system of claim 1, wherein the computing node is local to the near-infrared spectrometer.",
    "4. The system of claim 3, wherein the computing node and the near-infrared spectrometer are operatively coupled via a local area network.",
    "5. The system of claim 4, wherein the local area network comprises a wireless network.",
    "6. The system of claim 3, wherein the computing node and the near-infrared spectrometer are operatively coupled via a personal area network.",
    "7. The system of claim 3, wherein the computing node and the near-infrared spectrometer are integrated into a handheld device.",
    "8. The system of claim 1, wherein providing the at least one fixed-size vector to the trained classifier comprises sending the at least one fixed-size vector to a remote fabric validation server, and obtaining from the trained classifier the similarity score comprises receiving the similarity score from the fabric validation server.",
    "9. The system of claim 8, wherein said sending and receiving is performed via a wide area network.",
    "10. The system of claim 1, wherein extracting a plurality of features from the spectrum comprises noise reduction.",
    "11. The system of claim 1, wherein the method further comprises: receiving wavelength, intensity, and/or reflectance from the near-infrared spectrometer, and wherein extracting the plurality of features further comprises extracting features from the wavelength, intensity, and/or reflectance.",
    "12. The system of claim 1, wherein the trained classifier comprises an artificial neural network.",
    "13. The system of claim 1, wherein extracting the plurality of features comprises principal component analysis.",
    "14. The system of claim 1, wherein extracting the plurality of features comprises applying an artificial neural network.",
    "15. The system of claim 14, wherein the artificial neural network comprises at least one convolutional layer.",
    "16. The system of claim 15, wherein the artificial neural network comprises a plurality of 1D convolutional layers.",
    "17. The system of claim 1, wherein the method further comprises obtaining from the trained classifier a fabric material composition, weave type, thread count, yarn thickness, and/or color.",
    "18. The system of claim 1, wherein obtaining the similarity score comprises providing at least one fixed-size vector extracted from a reference sample to the trained classifier.",
    "19. The system of claim 1, the method further comprising providing the similarity score to a user.",
    "20. The system of claim 17, further comprising providing the fabric material composition, weave type, thread count, yarn thickness, and/or color to a user.",
    "21. A method comprising: receiving a near-infrared absorption spectrum of a fabric sample from a near-infrared spectrometer; extracting a plurality of features from the spectrum; generating at least one fixed-size vector from the plurality of features; providing the at least one fixed-size vector to a trained classifier; obtaining from the trained classifier a similarity score indicative of the similarity of the fabric sample to a reference fabric sample.",
    "22. The method of claim 21, wherein the near-infrared spectrometer comprises a hand-held spectrometer.",
    "23. The method of claim 21, wherein providing the at least one fixed-size vector to the trained classifier comprises sending the at least one fixed-size vector to a remote fabric validation server, and obtaining from the trained classifier the similarity score comprises receiving the similarity score from the fabric validation server.",
    "24. The method of claim 23, wherein said sending and receiving is performed via a wide area network.",
    "25. The method of claim 21, wherein extracting a plurality of features from the spectrum comprises noise reduction.",
    "26. The method of claim 21, further comprising: receiving wavelength, intensity, and/or reflectance from the near-infrared spectrometer, and wherein extracting the plurality of features further comprises extracting features from the wavelength, intensity, and/or reflectance.",
    "27. The method of claim 21, wherein the trained classifier comprises an artificial neural network.",
    "28. The method of claim 21, wherein extracting the plurality of features comprises principal component analysis.",
    "29. The method of claim 21, wherein extracting the plurality of features comprises applying an artificial neural network.",
    "30. The method of claim 29, wherein the artificial neural network comprises at least one convolutional layer.",
    "31. The method of claim 30, wherein the artificial neural network comprises a plurality of 1D convolutional layers.",
    "32. The method of claim 21, further comprising obtaining from the trained classifier a fabric material composition, weave type, thread count, yarn thickness, and/or color.",
    "33. The method of claim 21, wherein obtaining the similarity score comprises providing at least one fixed-size vector extracted from a reference sample to the trained classifier.",
    "34. The method of claim 21, the method further comprising providing the similarity score to a user.",
    "35. The method of claim 32, further comprising providing the fabric material composition, weave type, thread count, yarn thickness, and/or color to a user.",
    "36. A computer program product for fabric validation, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform a method comprising: receiving a near-infrared absorption spectrum of a fabric sample from a near-infrared spectrometer; extracting a plurality of features from the spectrum; generating at least one fixed-size vector from the plurality of features; providing the at least one fixed-size vector to a trained classifier; obtaining from the trained classifier a similarity score indicative of the similarity of the fabric sample to a reference fabric sample."
  ],
  "description_excerpt": "Embodiments of the present disclosure relate to fabric validation, and more specifically, to fabric validation using spectral measurement.\n\nAccording to embodiments of the present disclosure, methods of and computer program products for fabric validation are provided. In various embodiments, a near-infrared absorption spectrum of a fabric sample is received from a near-infrared spectrometer. A plurality of features is extracted from the spectrum. The plurality of features is provided to a trained classifier. The trained classifier provides a similarity score indicative of the similarity of the fabric sample to a reference fabric sample.\n\nIn various embodiments, the near-infrared spectrometer comprises a hand-held spectrometer. In various embodiments, providing the plurality of features to the trained classifier comprises sending the plurality of features to a remote fabric validation server, and obtaining from the trained classifier the similarity score comprises receiving the similarity score from the fabric validation server. In various embodiments, said sending and receiving is performed via a wide area network.\n\nIn various embodiments, extracting a plurality of features from the spectrum comprises noise reduction.\n\nIn various embodiments, wavelength, intensity, and/or reflectance are received from the near-infrared spectrometer, and extracting the plurality of features further comprises extracting features from the wavelength, intensity, and/or reflectance.\n\nIn various embodiments, the trained classifier comprises an artificial neural network.",
  "cpc": [
    "G01N 21/3563",
    "G01J 3/0264",
    "G01J 3/0272",
    "G01J 3/0291",
    "G01J 3/0297",
    "G01J 3/28",
    "G01J 3/42",
    "G01J 3/462",
    "G01N 21/359",
    "G01N 2201/0221",
    "G01N 2201/129",
    "G01N 2201/1296",
    "G01N 33/367",
    "G06F 2218/12",
    "G06K 9/00536",
    "G06N 3/0464",
    "G06N 3/08",
    "G06N 3/09"
  ],
  "ipc": [
    "G01J 3/02",
    "G01N 21/3563",
    "G01N 21/359",
    "G01J 3/28",
    "G01J 3/46",
    "G01N 33/36",
    "G06K 9/00",
    "G06N 3/08"
  ],
  "assignees": [
    "Inspectorio Inc"
  ],
  "inventors": [
    "Binh Thanh Nguyen",
    "Han Ky Cao",
    "Cuong Van Nguyen",
    "Carlos Moncayo"
  ],
  "filing_date": "2019-10-09",
  "publication_date": "2021-07-20",
  "grant_date": "2021-07-20",
  "priority_date": "2019-03-29",
  "application_number": "US-201916596997-A",
  "family_id": "72608200",
  "cited_by_count": 6,
  "citations": [
    "US6272479B1",
    "US7617163B2",
    "US7071469B2",
    "US20040119972A1",
    "US20100036795A1",
    "US8081304B2",
    "US8452716B2",
    "US20100290032A1",
    "US20170032285A1",
    "US10307795B2",
    "US20200320769A1",
    "US10936921B2",
    "US20200249085A1",
    "WO2020109170A1",
    "US20200042822A1"
  ]
}

Record 1,560 of 8,000 in Patents full text (MLC-0201). Request the full dataset.