Examination apparatus, examination method, recording medium storing an examination program, learning apparatus, learning method, and recording medium storing a learning program

Patent metadata

Field	Value
Publication number	US-11244443-B2
Application number	US-201916524162-A
Country	US
Kind code	B2
Filing date	Jul 28, 2019
Priority date	Jul 28, 2019
Publication date	Feb 8, 2022
Grant date	Feb 8, 2022

Abstract

Provided is an examination apparatus including a target image acquiring section that acquires a target image obtained by capturing an examination target; a target image masking section that masks a portion of the target image; a masked region predicting section that predicts an image of a masked region that is masked in the target image; a reproduced image generating section that generates a reproduced image using a plurality of predicted images predicted respectively for the plurality of masked regions; and a difference detecting section that detects a difference between the target image and the reproduced image.

First claim

What is claimed is:

1. An examination apparatus comprising:
a target image acquiring section that acquires a target image obtained by capturing an examination target image;
a target image masking section that masks a portion of the target image, the target image masking section being configured to divide the target image into a plurality of cells, each cell corresponding to a predetermined region that is to be masked, the target image masking section being configured to generate a plurality of masked images by sequentially masking each of the plurality of cells to generate a corresponding one of the plurality of masked images, such that each one of the plurality of masked images corresponds to a respective one of the plurality of cells and each one of the plurality of masked images has a predetermined masked region defined by the respective one of the plurality of cells;
a masked region predicting section that predicts a predicted image of the masked region in each one of the plurality of masked images to generate a plurality of predicted images predicted respectively for the masked region in each of the plurality of masked images;
a reproduced image generating section that generates a reproduced image using the plurality of predicted images by placing each of the plurality of predicted images in the same position in the reproduced image as the position of a corresponding one of the plurality of cells in the target image; and
a difference detecting section that detects a difference between the target image and the reproduced image.

2. The examination apparatus according to claim 1 , wherein the difference detecting section compares the target image to the reproduced image in every predetermined region, to calculate a degree of the difference in every predetermined region.

3. The examination apparatus according to claim 2 , further comprising: a judging section that judges the examination target to be unacceptable if the degree of difference does not satisfy a predetermined quality standard.

4. The examination apparatus according to claim 3 , wherein the judging section judges the examination target to be unacceptable if a largest degree of difference, among the degrees of difference of every predetermined region, exceeds a predetermined threshold value.

5. The examination apparatus according to claim 3 , wherein if the judging section judges the examination target to be unacceptable, the judging section predicts an electrical characteristic of the examination target from the target image obtained by capturing the examination target that was judged to be unacceptable, and confirms that the examination target is unacceptable if the electrical characteristic does not satisfy a predetermined quality standard.

6. The examination apparatus according to claim 3 , wherein if the judging section judges the examination target to be unacceptable, the judging section predicts an electrical characteristic of the examination target from the target image obtained by capturing the examination target that was judged to be unacceptable, and determines that the examination target is acceptable if the electrical characteristic satisfies a predetermined quality standard.

7. The examination apparatus according to claim 2 , wherein the difference detecting section outputs a detection map in which a display attribute differs in every predetermined region, according to the degree of difference.

8. The examination apparatus according to claim 2 , wherein the difference detecting section calculates the degree of difference based on a Euclidian distance between the target image and the reproduced image.

9. The examination apparatus according to claim 1 , wherein the target image acquiring section acquires an image obtained by performing a grayscale conversion on the captured image of the examination target, as the target image.

10. The examination apparatus according to claim 1 , wherein the target image acquiring section acquires an image obtained by performing object detection on the examination target in the captured image of the examination target to narrow a target region, as the target image.

11. An examination method comprising:
acquiring a target image obtained by capturing an examination target image;
sequentially masking a plurality of predetermined regions of the target image to generate a plurality of masked images each having a masked region corresponding to a respective one of the plurality of predetermined regions;
predicting a predicted image of the masked region of each of the plurality of masked images to generated a plurality of predicted images;
generating a reproduced image using the plurality of predicted images by placing each of the plurality of predicted images in the same position in the reproduced image as the position of the respective one of the plurality of predetermined regions in the target image; and
detecting a difference between the target image and the reproduced image.

12. A non-transitory computer-readable medium storing thereon an examination program that, when executed by a computer, causes the computer to function as:
a target image acquiring section that acquires a target image obtained by capturing an examination target image;
a target image masking section that masks a portion of the target image, the target image masking section being configured to divide the target image into a plurality of cells, each cell corresponding to a predetermined region that is to be masked, the target image masking section being configured to generate a plurality of masked images by sequentially masking each of the plurality of cells to generate a corresponding one of the plurality of masked images, such that each one of the plurality of masked images corresponds to a respective one of the plurality of cells and each one of the plurality of masked images has a predetermined masked region defined by the respective one of the plurality of cells;
a masked region predicting section that predicts a predicted image of the masked region in each one of the plurality of masked images to generate a plurality of predicted images predicted respectively for the masked region in each of the plurality of masked images;
a reproduced image generating section that generates a reproduced image using the plurality of predicted images by placing each of the plurality of predicted images in the same position in the reproduced image as the position of a corresponding one of the plurality of cells in the target image; and
a difference detecting section that detects a difference between the target image and the reproduced image.

13. The examination apparatus according to claim 1 , wherein the plurality of cells are of uniform size and shape.

14. The examination method according to claim 11 , wherein the plurality of predetermined regions are of uniform size and shape.

15. The non-transitory computer-readable medium according to claim 12 , wherein the plurality of cells are of uniform size and shape.

Assignees

• Advantest Corp

Inventors

• Ikeda Kosuke
• Leventhal Ira
• Schaub Keith

Classifications

• G06V10/7796

  based on specific statistical tests · CPC title

• G06V10/82

  using neural networks · CPC title

• G06V10/454

  Integrating the filters into a hierarchical structure, e.g. convolutional neural networks [CNN] · CPC title

• G06V10/25

  Determination of region of interest [ROI] or a volume of interest [VOI] · CPC title

• G06F18/214

  Generating training patterns; Bootstrap methods, e.g. bagging or boosting · CPC title