# US 10,500,842 B2 · Automated image sensor calibration

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

- Number: US 10,500,842 B2
- Document ID: US10500842B2
- Status: Granted
- Kind: B2
- Application number: 16/128,731
- Filed: 2018-09-12
- Published: 2019-03-14
- Granted: 2019-12-10
- Earliest claimed: 2017-09-12
- Assignees: Heidelberger Druckmaschinen AG
- Applicants: Heidelberger Druckmaschinen AG
- Inventors: Jan Krieger; Frank Muth
- Web page: https://patentgrep.com/patent/US10500842B2

## CPC

- B41J 2/04505: {aiming at correcting alignment}
- B41J 2/04586: {controlling heads of a type not covered by groups B41J2/04575 - B41J2/04585, or of an undefined type}
- B41J 2/2139: {Compensation for malfunctioning nozzles creating dot place or dot size errors (generating single droplets or particles on demand by pressure, e.g. electromechanical transducers B41J2/045)}
- B41J 2/2142: {Detection of malfunctioning nozzles (generating single droplets or particles on demand by pressure, e.g. electromechanical transducers B41J2/045, B41J2/05; jet deflection sensors B41J2/125; for cleaning purposes B41J2/16579)}
- G06K 15/027: {Test patterns and calibration (arrangements for controlling or analysing printing mechanisms of typewriters or selective printing units B41J29/393 takes precedence; colour correction using test pattern analysis in general H04N1/6033)}
- H04N 1/6033: {using test pattern analysis (H04N1/6055 takes precedence)}
- B41J 2/14: Structure thereof {only for on-demand ink jet heads}
- B41J: TYPEWRITERS; SELECTIVE PRINTING MECHANISMS, i.e. MECHANISMS PRINTING OTHERWISE THAN FROM A FORME; CORRECTION OF TYPOGRAPHICAL ERRORS
- G06K: GRAPHICAL DATA READING (image or video recognition or understanding G06V); PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION

## Abstract

A method calibrates image sensors and adjusts print heads in an inkjet printing machine via a computer. The method includes printing a test print chart, recording the printed test print chart via an image sensor to create a digital image, determining a coordinate transformation between the coordinates on the test print chart and the coordinates of the image sensor and the print heads, implementing a computer-assisted evaluation process of the digital image to determine the coordinates on the test print chart, deriving correction values for print head adjustment and camera calibration from results of the coordinate transformation and/or the comparison thereof with the determined measured values, and implementing the print head adjustment and the camera calibration by the computer. The test print chart has filled circular discs. The filled circular discs have a minimum diameter that is selected for them to function even if defective printing nozzles occur.

## Claims

### Claim 1

A method for calibrating image sensors and adjusting print heads in an inkjet printing machine in an automated way via a computer, which comprises the steps of:
printing a test print chart containing filled circular discs functioning as calibration dots, the filled circular discs having a minimum diameter that is selected in a way for them to function in an optimal way even if defective printing nozzles occur;
recording the test print chart by means of at least one image sensor to create a digital image;
determining a coordinate transformation between coordinates on the test print chart and coordinates of the at least one image sensor and the print heads via the computer;
implementing a computer-assisted evaluation process of the digital image to determine the coordinates on the test print chart;
deriving correction values for print head adjustment and camera at least one image sensor calibration from results of the coordinate transformation and/or a comparison thereof with determined measured values of the test print chart by means of the computer;
implementing the print head adjustment and the at least one image sensor calibration by means of the computer and using the correction values; and
subdividing a printing substrate region to be printed and recorded by the at least one image sensor into image sensor regions corresponding to a number of the image sensors that are used, every one of the image sensor regions include a reference dot in a form of a filled circular disc printed centrally above the calibration dots, the reference dot being used for referencing within an individual image sensor region of the image sensor regions.

### Claim 2 (depends on 1)

The method according to claim 1, wherein for all color separations that are used, the test print chart containing the filled circular discs is printed, every print head printing at least two complete rows of at least three said calibration dots in a form of the filled circular discs, the filled circular discs not in a stitching range of the print head and the at least two complete rows distributed over a printing substrate in such a way that they span a largest possible area in a direction of paper transport.

### Claim 3 (depends on 1)

The method according to claim 1, wherein when the test print chart containing the filled circular discs is printed, every print head printing at least two complete blocks of rows of at least two said calibration dots in a form of the filled circular discs, the filled circular discs not located in a stitching region of the print head, and every block including at least one row of at least two said calibration dots for every color separation that is used and the rows of at least two said calibration dots distributed over a printing substrate in such a way that they span a largest possible area in a direction of paper transport.

### Claim 4 (depends on 1)

The method according to claim 1, which further comprises providing a camera system to record the printed test print chart by means of the at least one image sensor, the camera system containing multiple line scan cameras disposed next to one another, the line scan cameras recording the test print chart at a resolution of at least 40-60% of a print resolution.

### Claim 5 (depends on 1)

The method according to claim 1, wherein a computer-assisted evaluation process for determining the coordinates on the test print chart being done by detecting a center of the calibration dots and comprises the further following steps of:
converting a recorded test print chart into a high-contrast grayscale image;
cutting an image region including a circular disc out of an entire digital image;
performing at least one of:
implementing an edge detection process;
converting the image region into a binary image and filtering line artifacts out of the binary image;
adding a cross through a center of the circular disc to the binary image to create a binary mask;
widening the binary mask by dilatation;
determining parameters of a model of a print dot from the entire digital image, using the binary mask; and
checking results, namely whether the center and a radius of the circular disc are within an expected range.

### Claim 6 (depends on 5)

The method according to claim 5, wherein the coordinate transformation implemented by the computer comprises the further following steps of:
searching the reference dot in a currently evaluated region of the test print chart by means of a theoretically calculated coordinate transformation that is already present;
selecting a color channel with a highest contrast if a camera has multiple color channels;
refining an initial coordinate transformation using center coordinates of the reference dot;
determining positions of available calibration dots by cutting out image regions about respective calibration dots and implementing the evaluation process;
filtering out deviating calibration dots and errors;
determining parameters of a model of the coordinate transformation using remaining calibration dots if sufficient calibration dots remain after filtering;
determining a deviation of the calibration dots from their expected ideal position; and
combining deviations as results for the print heads.

### Claim 7 (depends on 5)

The method according to claim 5, wherein the coordinate transformations calculated for multiple color separations are used to determine register deviations in a form of positional deviations of individual print colors relative to one another and to correct them.

### Claim 8 (depends on 7)

The method according to claim 7, wherein calculation of the register deviations is implemented at various locations distributed over the test print chart to determine a register deviation average.