# US 2026/0094255 A1 · Method for detecting a defect in a via, detection device, storage medium and evaluation device

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0094255 A1
- Document ID: US20260094255A1
- Status: Published application
- Kind: A1
- Application number: 19/347,964
- Filed: 2025-10-02
- Published: 2026-04-02
- Earliest claimed: 2024-10-02
- Assignees: Siemens Healthineers AG
- Applicants: Siemens Healthineers AG
- Inventors: Benedikt Christopher VOGT-WOLZ; Jan WREGE
- Web page: https://patentgrep.com/patent/US20260094255A1

## CPC

- G06T 7/0004: {Industrial image inspection}
- G06T 7/11: Region-based segmentation
- G06T 2207/10056: Microscopic image
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T 2207/30148: Semiconductor; IC; Wafer
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A method for detecting at least one defect in at least one hollow via of a microelectronic component, the method comprising: acquiring, via an image acquisition device, image data of a hollow interior space of the via relating to at least a part of the visually visible range of the electromagnetic spectrum; supplying the image data to an evaluation device; and evaluating the image data to identify the at least one defect.

## Claims

### Claim 1

A method for detecting at least one defect in at least one hollow via of a microelectronic component, the method comprising:
acquiring, via an image acquisition device, image data of a hollow interior space of the at least one hollow via relating to at least a part of a visually visible range of an electromagnetic spectrum;
supplying the image data to an evaluation device; and
evaluating, via the evaluation device, the image data to identify the at least one defect.

### Claim 2 (depends on 1)

The method as claimed in claim 1, wherein at least one of
the microelectronic component includes a substrate,
the at least one hollow via is formed by a recess in the substrate, or
the hollow interior space of the at least one hollow via is bounded by a layer composed of a metal that is applied via chemical or physical deposition.

### Claim 3 (depends on 1)

The method as claimed in claim 1, wherein
the image acquisition device comprises at least one light source configured to generate light, and
the image data is acquired while the at least one hollow via is illuminated by the light generated by the at least one light source.

### Claim 4 (depends on 3)

The method as claimed in claim 3, wherein the light is polarized.

### Claim 5 (depends on 1)

The method as claimed in claim 1, wherein a line of sight of the image acquisition device stands at least substantially perpendicularly on a base of the at least one hollow via during acquisition of the image data.

### Claim 6 (depends on 1)

The method as claimed in claim 1, wherein
the image acquisition device comprises a microscope and a camera, and
the image data relates to a visualization of the at least one hollow via magnified via the microscope.

### Claim 7 (depends on 6)

The method as claimed in claim 6, wherein acquisition of the image data is performed in the course of at least one of (i) a brightfield microscopy or a darkfield microscopy inspection, or (ii) an operating mode for implementing a differential interference contrast or a phase contrast.

### Claim 8 (depends on 1)

The method as claimed in claim 1, further comprising:
segmenting the image data via the evaluation device, wherein
resulting image segments are assigned to at least one of defects or defined regions of the at least one hollow via.

### Claim 9 (depends on 1)

The method as claimed in claim 1, further comprising:
performing, via the evaluation device, a classification of a type of a respective defect for a detected defect or for at least one detected defect.

### Claim 10 (depends on 9)

The method as claimed in claim 9, further comprising:
indicating, based on the classification, that the respective defect is at least one of a delamination or a bulge.

### Claim 11 (depends on 1)

The method as claimed in claim 1, wherein the evaluating evaluates the image data to identify the at least one defect using an artificial intelligence application.

### Claim 12 (depends on 1)

A detection device for performing the method as claimed in claim 1, the detection device comprising:
the image acquisition device configured to acquire image data of the hollow interior space of the at least one hollow via; and
the evaluation device configured to receive and evaluate the image data to identify the at least one defect.

### Claim 13 (depends on 1)

A non-transitory computer-readable storage medium comprising instructions that, when executed by a processing device, cause the processing device to perform the method as claimed in claim 1.

### Claim 14 (depends on 13)

An evaluation device comprising:
the non-transitory computer-readable storage medium as claimed in claim 13; and
a processing device configured to execute the instructions.

### Claim 15 (depends on 1)

A computer-implemented method for generating a trained model usable in the method as claimed in claim 1 in the course of evaluating the image data to identify the at least one defect, wherein the computer-implemented method comprises:
specifying at least one training input dataset;
specifying a training result assigned to the at least one training input dataset; and
training a model based on the at least one training input dataset and the training result, to obtain the trained model.

### Claim 16 (depends on 2)

The method as claimed in claim 2, wherein at least one of
the substrate is comprised of silicon,
the at least one hollow via is formed by etching the substrate, or
the metal is copper.

### Claim 17 (depends on 4)

The method as claimed in claim 4, wherein the light is polarized circularly.

### Claim 18 (depends on 8)

The method as claimed in claim 8, wherein the defined regions of the at least one hollow via include a base or a sidewall of the at least one hollow via.

### Claim 19 (depends on 11)

The method as claimed in claim 11, wherein the artificial intelligence application is a trained model generated via a machine learning process.

### Claim 20 (depends on 3)

The method as claimed in claim 3, further comprising:
performing, via the evaluation device, a classification of a type of a respective defect for a detected defect or for at least one detected defect.
