Apparatus for detecting defects in printed images, and method for detecting defects therein.

JP7877717B2Active Publication Date: 2026-06-23SEIKO EPSON CORP

Patent Information

Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
SEIKO EPSON CORP
Filing Date
2022-03-07
Publication Date
2026-06-23

Smart Images

  • Figure 0007877717000001
    Figure 0007877717000001
  • Figure 0007877717000002
    Figure 0007877717000002
  • Figure 0007877717000003
    Figure 0007877717000003
Patent Text Reader

Abstract

To intelligibly display what kind of defect has occurred.SOLUTION: A defect discrimination device for printed image includes: an inspection image acquisition section configured to acquire an inspection image by imaging a print medium on which an image is printed, the image corresponding to a reference image that is a reference for an image in which a defect is to be detected; a defect detection section configured to detect a state of the defect, the state including a defect type of the defect included in the printed inspection image; a display section configured to display a display image corresponding to the inspection image; and an inspection result display section configured to display the detected defect together with the defect type of the defect, in a display mode according to the state of the defect.SELECTED DRAWING: Figure 2
Need to check novelty before this filing date? Find Prior Art

Claims

1. A device for detecting defects in printed images, An inspection image acquisition unit acquires an inspection image by capturing a printed medium on which an image corresponding to a reference image that will serve as the basis for detecting defects in the image, A difference extraction unit that extracts a difference image between the reference image and the inspection image, A defect detection unit that uses the difference image to detect the similarity of each type of defect in the image contained in the inspection image, An inspection result display unit that, when a defect type set by the user has a similarity score above a predetermined threshold and is detected by the defect detection unit, displays the state of the defect, including the similarity score to the defect type, together with the defect type of the defect, in a display manner appropriate to the state of the defect. A defect detection device equipped with the following features.

2. A defect detection device according to claim 1, The defect detection unit identifies the location of the defect on the inspection image along with the state of the defect. The inspection result display unit displays the defect at a position corresponding to the location where the defect was identified on the inspection image. Defect detection device.

3. A defect detection device according to claim 1 or claim 2, A defect determination device in which the state of the defect includes, in addition to the degree of similarity to the defect type, at least one of the defect type, the location of the defect, the size of the defect, and the shape of the defect.

4. A defect detection device according to any one of claims 1 to 3, The inspection result display unit is a defect discrimination device that displays the defect in a display manner that includes at least one of color coding, shading, inversion, flashing, or framing.

5. A defect detection device according to any one of claims 1 to 4, The aforementioned inspection result display unit is: The aforementioned display mode can be displayed using multiple types of display methods. The detected defects are displayed in a list according to the defect type. It is possible to set whether to display each of the defect types in the list, depending on the type of defect according to its condition. Defect detection device.

6. A defect detection device according to claim 5, The inspection result display unit is a defect discrimination device that sets the display mode for each type of defect.

7. A defect discrimination device according to Claim 1, The defect detection unit includes a similarity calculation unit that uses a machine learning model generated by linking the defect and the defect type to infer the defect from the difference image and calculate the similarity to the defect type, thereby providing a defect discrimination device.

8. A method for detecting defects in printed images, A process of acquiring an inspection image by imaging a printed medium on which an image corresponding to a reference image that will serve as the basis for detecting defects in the image, A step of extracting a difference image between the reference image and the inspection image, Using the difference image, a step is to detect the similarity of each type of defect in the image contained in the inspection image, which has been prepared in advance. A step in which, when a defect type set by the user has a similarity of a predetermined threshold or higher, is detected in the detection step, the state of the defect, including its similarity to the defect type, is displayed together with the defect type in a display manner appropriate to the state of the defect. A defect detection method comprising the following features.