# US 2025/0061558 A1 · Test device and operating method thereof

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

- Number: US 2025/0061558 A1
- Document ID: US20250061558A1
- Status: Published application
- Kind: A1
- Application number: 18/805,724
- Filed: 2024-08-15
- Published: 2025-02-20
- Earliest claimed: 2023-08-18
- Assignees: Samsung Electronics Co., Ltd.
- Applicants: Samsung Electronics Co., Ltd.
- Inventors: Yubin LEE; Dayun LEE; Kundong KIM; Yongkoo PARK
- Web page: https://patentgrep.com/patent/US20250061558A1

## CPC

- G06T 7/0004: {Industrial image inspection}
- G06T 7/001: {using an image reference approach}
- G06V 10/764: using classification, e.g. of video objects
- G06V 10/82: using neural networks
- G06T 2207/20081: Training; Learning
- G06T 2207/20084: Artificial neural networks [ANN]
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A system for testing whether or not an image sensor included in an image frame is defective includes including a processor configured to execute machine-readable instructions that, when executed by the processor, cause the test device to generate a test image by performing one or more preprocessing operations on a raw image output from an image sensor, and to classify an image pattern of the test image as any one of patterns included in a first data set by using a first deep learning neural network trained based on the first data set and to determine, based on a classifying result, whether or not the image sensor is defective. The first data set comprises a data set for each pattern, classified to correspond to each of one or more defect patterns and to a normal pattern of an image.

## Claims

### Claim 1

A test device comprising:
a processor configured to execute instructions that, when executed by the processor, cause the test device to
generate a test image by performing one or more preprocessing operations on a raw image output from an image sensor, and
to classify an image pattern of the test image as any one of patterns included in a first data set by using a first deep learning neural network trained based on the first data set and to determine, based on a classifying result, whether or not the image sensor is defective,
wherein the first data set comprises a data set for each pattern, classified to correspond to each of one or more defect patterns and to a normal pattern of an image.

### Claim 2 (depends on 1)

The test device of claim 1, wherein the test device is further caused to,
calculate, based on a training result of the first deep learning neural network, a matching score between the image pattern and the normal pattern and a matching score between the image pattern and each of the one or more defect patterns, and
classify the image pattern of the test image based on a highest matching score from among the calculated scores.

### Claim 3 (depends on 1)

The test device of claim 1, wherein the test device is further caused to,
determine that the image sensor is a good product in response to classifying the image pattern as the normal pattern, and
determine that the image sensor is a defective product of a type corresponding to the image pattern in response to classifying the image pattern as any one of the one or more defect patterns.

### Claim 4 (depends on 1)

The test device of claim 1, wherein the one or more defect patterns comprise a spot pattern, a diagonal pattern, and an error pattern.

### Claim 5 (depends on 1)

The test device of claim 1, wherein the one or more preprocessing operations comprise one or more of a calibration operation, a fixed-point noise reduction (FPNR) operation, and a filtering operation.

### Claim 6 (depends on 1)

The test device of claim 1, wherein the test device is further caused to determine a defect level of the test image by using a second deep learning neural network trained based on a second data set, wherein the defect level is determined based on a visibility of entire defect patterns included in the test image.

### Claim 7 (depends on 6)

The test device of claim 6, wherein the second data set comprises a data set for each defect level classified according to an intensity of the visibility of the entire defect patterns, wherein training the second deep learning neural network includes using the intensity of the visibility of the entire defect patterns.

### Claim 8 (depends on 6)

The test device of claim 6, wherein the test device is further caused to,
calculate, based on a training result of the second deep learning neural network, a matching score between the entire defect patterns included in the test image and the data set for each defect level, and
determine the defect level of the test image as a level having a highest matching score from among the calculated scores.

### Claim 9 (depends on 6)

The test device of claim 6, wherein the first deep learning neural network is trained based on a first Huber-loss function, and the second deep learning neural network is trained based on a second Huber-loss function.

### Claim 10

An operating method of a test device, the operating method comprising:
generating a test image by performing one or more preprocessing operations on a raw image output from an image sensor;
classifying an image pattern of the test image as any one of patterns included in a first data set by using a first deep learning neural network trained based on the first data set; and
determining, based on a result of the classifying, whether or not the image sensor is defective,
wherein the first data set comprises a data set for each pattern classified to correspond to each of one or more defect patterns and to a normal pattern of an image.

### Claim 11 (depends on 10)

The operating method of claim 10, wherein the classifying of the image pattern comprises:
calculating, based on a training result of the first deep learning neural network, a matching score between the image pattern and the normal pattern and a matching score between the image pattern and each of the one or more defect patterns; and
classifying the image pattern of the test image based on a highest matching score from among the calculated scores.

### Claim 12 (depends on 10)

The operating method of claim 10, wherein the determining of whether or not the image sensor is defective comprises:
determining that the image sensor is a good product in response to the image pattern being classified as the normal pattern; and
determining that the image sensor is a defective product of a type corresponding to the image pattern in response to the image pattern is classified as any one of the one or more defect patterns.

### Claim 13 (depends on 10)

The operating method of claim 10, wherein the one or more defect patterns comprise a spot pattern, a diagonal pattern, and an error pattern.

### Claim 14 (depends on 10)

The operating method of claim 10, wherein the one or more preprocessing operations comprise one or more of a calibration operation, a fixed-point noise reduction (FPNR) operation, and a filtering operation.

### Claim 15 (depends on 10)

The operating method of claim 10, further comprising:
determining a defect level of the test image by using a second deep learning neural network trained based on a second data set,
wherein the defect level is determined based on a visibility of entire defect patterns included in the test image.

### Claim 16 (depends on 15)

The operating method of claim 15, wherein the second data set comprises a data set for each defect level, classified according to an intensity of the visibility of the entire defect patterns, wherein training the second deep learning neural network includes using an intensity of the visibility of the entire defect patterns.

### Claim 17 (depends on 15)

The operating method of claim 15, wherein the determining of whether or not the image sensor is defective further comprises:
calculating, based on a training result of the second deep learning neural network, a matching score between the entire defect patterns included in the test image and the data set for each defect level; and
determining the defect level of the test image based on a highest matching score from among the calculated scores.

### Claim 18 (depends on 15)

The operating method of claim 15, wherein the first deep learning neural network is trained based on a first Huber-loss function, and the second deep learning neural network is trained based on a second Huber-loss function.

### Claim 19

A system for testing whether or not an image sensor is defective, the system comprising:
a memory configured to store one or more instructions; and
a processor configured to execute the one or more instructions stored in the memory to,
generate a test image by performing one or more preprocessing operations on a raw image received from an image sensor,
detect an image pattern of the test image by using a first deep learning neural network trained based on a first data set,
determine, based on a detecting result regarding the image pattern, whether or not the image sensor is defective, and
determine a defect level of the test image by using a second deep learning neural network trained based on a second data set.

### Claim 20 (depends on 19)

The system of claim 19, wherein
the first data set comprises a data set for each pattern, classified to correspond to a normal pattern of an image and each of one or more defect patterns, and
the second data set comprises a data set for each defect level, -classified according to an intensity of a visibility of entire defect patterns.
