# US 2013/0240628 A1 · Barcode Recognition Using Data-Driven Classifier

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

- Number: US 2013/0240628 A1
- Document ID: US20130240628A1
- Status: Published application
- Kind: A1
- Application number: 13/658,765
- Filed: 2012-10-23
- Published: 2013-09-19
- Earliest claimed: 2010-09-30
- Assignees: Apple Inc.
- Applicants: Apple Inc.
- Inventors: Rudolph van der Merwe; Samuel Gordon Noble; Ralph Brunner
- Web page: https://patentgrep.com/patent/US20130240628A1

## CPC

- G06K 7/1447: {extracting optical codes from image or text carrying said optical code}
- G06K 7/10811: {Focalisation}
- G06K 7/1443: {locating of the code in an image}
- G06K: GRAPHICAL DATA READING (image or video recognition or understanding G06V); PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS

## Abstract

A barcode decoding system and method are disclosed that use a data-driven classifier for transforming a potentially degraded barcode signal into a digit sequence. The disclosed implementations are robust to signal degradation through incorporation of a noise model into the classifier construction phase. The run-time computational cost is low, allowing for efficient implementations on portable devices. Implementations are disclosed for intelligent preview scaling, barcode-aware autofocus augmentation and multi-scale signal feature extraction.

## Claims

### Claim 1

A method comprising:
capturing a digital image of an object using a sensor of a device; and
scaling the image to lie within a predefined image view area of a user interface presented on a display screen of the device, while a distance is maintained between the object and the sensor,
where the method is performed by one or more processors.

### Claim 2 (depends on 1)

The method of claim 1, wherein the object is a barcode.

### Claim 3 (depends on 1)

The method of claim 1, wherein the scaling further comprises:
determining, using a geometric model, mapping information that maps a physical, real-world coordinate frame onto a sensor coordinate frame.

### Claim 4 (depends on 3)

The method of claim 3, further comprising:
constraining the size of the predefined image view area of the user interface using the mapping information.

### Claim 5 (depends on 4)

The method of claim 4, where constraining the size of the predefined image view area of the user interface is based in part on the optical magnification of the object or constraints of the sensor.

### Claim 6 (depends on 3)

The method of claim 3, where the geometric model is a trigonometric model of a pinhole camera.

### Claim 7 (depends on 6)

The method of claim 6, where the object has a length h and the object's image on the camera sensor has a dimension of length h′, and the pinhole camera model relates the distance d of the object from a camera lens of the camera system to the distance of the camera sensor from the camera lens d′, and the relationship is given by h/d=h′/d′.

### Claim 8 (depends on 3)

The method of claim 3, where the geometric model accounts for one of lens distortion, sensor characteristics and photographic effects.

### Claim 9 (depends on 8)

The method of claim 8, further comprising:
estimating geometric model parameters using a calibration process and an image of a calibration target taken from a number of positions of the sensor relative to the calibration target.

### Claim 10 (depends on 1)

The method of claim 1, where the scaling is performed using a scale factor that is computed relative to a target guide and that is contained entirely within the predefined image view area, and where the scale factor scales the image to substantially fill the predefined image view area while preserving an aspect ratio of the image.

### Claim 11

A method comprising:
capturing a digital image of an object using a sensor of a device;
detecting that the object is out of focus using a focus metric; and
automatically refocusing the image at a target point in the image in response to the detecting,
where the method is performed by one or more processors.

### Claim 12 (depends on 11)

The method of claim 11, further comprising:
determining the target point based on one of a user interface displaying the digital image, user input and an automatic object location process.

### Claim 13 (depends on 11)

The method of claim 11, where the object is a barcode.

### Claim 14 (depends on 11)

The method of claim 11, where the focus metric is determined by:
constructing a histogram distribution from differentiation of an image scan signal;
calculating an amount of probability mass in tails of the histogram distribution;
comparing the calculated amount with a predetermined threshold value; and
determining the focus metric based on the comparing.

### Claim 15 (depends on 14)

The method of claim 14, where constructing a histogram distribution further comprises:
calculating a first order difference signal from a normalized linear scan across the sensor; and
assigning values of the difference signal into bins of the histogram distribution.

### Claim 16 (depends on 14)

The method of claim 14, where the threshold value is an amount of normalized probability mass needed for the image to be in-focus.

### Claim 17

A method comprising:
capturing a digital image of an object on a sensor of a device;
generating an image signal from the sensor;
calculating an envelope of the image signal at different spatial scales; and
filtering the envelope to extract object features,
where the method is performed by one or more processors.

### Claim 18 (depends on 17)

The method of claim 17, where the object is a barcode and the object features are barcode symbol features.

### Claim 19 (depends on 17)

The method of claim 17, further comprising:
training a learning based classifier using the object features.

### Claim 20 (depends on 19)

The method of claim 19, where the learning based classifier is a neural network configured to provide a decoding quality metric.

### Claim 21 (depends on 20)

The method of claim 20, where the decoding quality metric is used in the calculating of the envelope of the image signal at the differential spatial scales.

### Claim 22

A system comprising:
one or more processors;
memory configured for storing instructions, which, when executed by the one or more processors, causes the one or more processors to perform operations comprising:
capturing a digital image of an object using a sensor of a device; and
scaling the image to lie within a predefined image view area of a user interface presented on a display screen of the device, while a distance is maintained between the object and the sensor.

### Claim 23 (depends on 22)

The system of claim 22, wherein the object is a barcode.

### Claim 24 (depends on 22)

The system of claim 22, wherein the scaling further comprises:
determining, using a geometric model, mapping information that maps a physical, real-world coordinate frame onto a sensor coordinate frame.

### Claim 25 (depends on 24)

The system of claim 24, further comprising:
constraining the size of the predefined image view area of the user interface using the mapping information.

### Claim 26 (depends on 25)

The system of claim 25, where constraining the size of the predefined image view area of the user interface is based in part on the optical magnification of the object or constraints of the sensor.

### Claim 27 (depends on 24)

The system of claim 24, where the geometric model is a trigonometric model of a pinhole camera.

### Claim 28 (depends on 27)

The system of claim 27, where the object has a length h and the object's image on the camera sensor has a dimension of length h′, and the pinhole camera model relates the distance d of the object from a camera lens of the camera system to the distance of the camera sensor from the camera lens d′, and the relationship is given by h/d=h′/d′.

### Claim 29 (depends on 24)

The system of claim 24, where the geometric model accounts for one of lens distortion, sensor characteristics and photographic effects.

### Claim 30 (depends on 24)

The system of claim 24, further comprising:
estimating geometric model parameters using a calibration process and an image of a calibration target taken from a number of positions of the sensor relative to the calibration target.

### Claim 31 (depends on 22)

The system of claim 22, where the scaling is performed using a scale factor that is computed relative to a target guide and that is contained entirely within the predefined image view area, and where the scale factor scales the image to substantially fill the predefined image view area while preserving an aspect ratio of the image.

### Claim 32

A system comprising:
one or more processors;
memory configured for storing instructions, which, when executed by the one or more processors, causes the one or more processors to perform operations comprising:
capturing a digital image of an object using a sensor of a device;
detecting that the object is out of focus using a focus metric; and
automatically refocusing the image at a target point in the image in response to the detecting.

### Claim 33 (depends on 32)

The system of claim 32, further comprising:
determining the target point based on one of a user interface displaying the digital image, user input and an automatic object location process.

### Claim 34 (depends on 32)

The system of claim 32, where the object is a barcode.

### Claim 35 (depends on 32)

The system of claim 32, where the focus metric is determined by:
constructing a histogram distribution from differentiation of an image scan signal;
calculating an amount of probability mass in tails of the histogram distribution;
comparing the calculated amount with a predetermined threshold value; and
determining the focus metric based on the comparing.

### Claim 36 (depends on 35)

The system of claim 35, where constructing a histogram distribution further comprises:
calculating a first order difference signal from a normalized linear scan across the sensor; and
assigning values of the difference signal into bins of the histogram distribution.

### Claim 37 (depends on 35)

The system of claim 35, where the threshold value is an amount of normalized probability mass needed for the image to be in-focus.

### Claim 38

A system comprising:
one or more processors;
memory configured for storing instructions, which, when executed by the one or more processors, causes the one or more processors to perform operations comprising:
capturing a digital image of an object on a sensor of a device;
generating an image signal from the sensor;
calculating an envelope of the image signal at different spatial scales; and
filtering the envelope to extract object features.

### Claim 39 (depends on 38)

The system of claim 38, where the object is a barcode and the object features are barcode symbol features.

### Claim 40 (depends on 38)

The system of claim 38, further comprising:
training a learning based classifier using the object features.

### Claim 41 (depends on 40)

The system of claim 40, where the learning based classifier is a neural network configured to provide a decoding quality metric.

### Claim 42 (depends on 41)

The system of claim 41, where the decoding quality metric is used in the calculating of the envelope of the image signal at the differential spatial scales.
