Automated inspection system
US-2024420305-A1 · Dec 19, 2024 · US
US2021295485A1 · US · A1
| Field | Value |
|---|---|
| Publication number | US-2021295485-A1 |
| Application number | US-201616343306-A |
| Country | US |
| Kind code | A1 |
| Filing date | Dec 6, 2016 |
| Priority date | Dec 6, 2016 |
| Publication date | Sep 23, 2021 |
| Grant date | — |
A practical reading order for non-experts. Skip the full description unless you need deep technical detail.
What the patent document calls the invention.
A short plain-language summary of the technical disclosure.
Who owns or filed the patent and who is credited as inventor.
Filing, priority, publication, and grant dates set the timeline.
The legal scope of protection — read this for what is actually claimed.
Technology tags used to group this patent with similar filings.
Prior art links and similar publications in this corpus.
Official abstract text for this publication.
An inspection device includes: an analyzer to calculate a parameter representing a feature of image data of an object having no defect by performing dimensionality reduction on the image data, and perform dimensionality reduction on image data of an object to be inspected by using the parameter; a restorer to generate restored data obtained by restoring the image data of the object to be inspected subjected to the dimensionality reduction; a corrector to filter the restored data by using a filter for correcting an error between the restored data and the image data of the object to be inspected, thereby generating corrected restored data; a determiner to output a determination result indicating whether the object to be inspected is defective, based on a difference of each pixel between the image data of the object to be inspected and the corrected restored data; and an interface to output the determination result.
Opening claim text (preview).
1 . An inspection device comprising: an analyzer to calculate a parameter representing a feature of image data of an object having no defect by performing dimensionality reduction for reducing a dimensionality of data on the image data of the object having no defect, and perform dimensionality reduction on image data of an object to be inspected by using the parameter; a restorer to generate restored data obtained by restoring the image data of the object to be inspected subjected to the dimensionality reduction by the analyzer; a corrector to filter the restored data by using a filter for correcting an error between the restored data restored by the restorer and the image data of the object to be inspected, thereby generating corrected restored data; a determiner to output a determination result indicating whether the object to be inspected is defective, on a basis of a magnitude of a difference of each pixel between the image data of the object to be inspected and the corrected restored data; and an interface to output the determination result output by the determiner. 2 . The inspection device of claim 1 , further comprising a storage medium to store the parameter calculated by the analyzer, wherein the analyzer performs the dimensionality reduction on the image data of the object to be inspected by using the parameter stored in the storage medium. 3 . (canceled) 4 . The inspection device of claim 1 , wherein the filter is a Wiener filter. 5 . The inspection device of claim 1 , wherein when at least one of the differences is not less than a threshold value, the determiner outputs a determination result indicating that the object to be inspected is defective. 6 . (canceled) 7 . The inspection device of claim 1 , wherein the dimensionality reduction uses principal component analysis. 8 . The inspection device of claim 1 , wherein the dimensionality reduction uses an autoencoder using a neural network. 9 . The inspection device of claim 8 , wherein the restorer generates the restored data by using the parameter. 10 . The inspection device of claim 5 , wherein the determiner calculates a bounding box that is a rectangle circumscribing a region where the difference is not less than the threshold value and a determination result indicating that the object to be inspected is defective is output; and the interface outputs, as the determination result, the bounding box calculated by the determiner. 11 . The inspection device of claim 10 , wherein the interface performs the output so that a defective portion of the image data of the object to be inspected determined on a basis of the bounding box is indicated by a dotted line superimposed on the image data of the object to be inspected. 12 . The inspection device of claim 10 , wherein the interface outputs difference composite image data obtained by combining image data of the difference with the image data of the object to be inspected. 13 . The inspection device of claim 1 , wherein when the determiner outputs a determination result indicating that the object to be inspected is defective, the interface performs the output so that a message for informing that a defect has been detected is displayed. 14 . The inspection device of claim 1 , further comprising a display to display the determination result output by the interface. 15 . An inspection method comprising: calculating a parameter representing a feature of image data of an object having no defect by performing dimensionality reduction for reducing a dimensionality of data on the image data of the object having no defect, and performing dimensionality reduction on image data of an object to be inspected by using the parameter stored in a storage medium; generating restored data obtained by restoring the image data of the object to be inspected subjected to the dimensionality reduction; filtering the restored data by using a filter for correcting an error between the restored data and the image data of the object to be inspected, thereby generating corrected restored data; outputting a determination result indicating whether the object to be inspected is defective, on a basis of a magnitude of a difference of each pixel between the image data of the object to be inspected and the corrected restored data; and outputting the determination result.
using an image reference approach · CPC title
Patterns showing highly reflecting parts, e.g. metallic elements · CPC title
for PCB's · CPC title
based on image processing techniques · CPC title
using local operators · CPC title
Related publications grouped by family.
Answers are generated from the same data shown on this page.