# US 2023/0419529 A1 · Method and apparatus for acquiring master data of an object

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

- Number: US 2023/0419529 A1
- Document ID: US20230419529A1
- Status: Published application
- Kind: A1
- Application number: 18/201,330
- Filed: 2023-05-24
- Published: 2023-12-28
- Earliest claimed: 2022-05-25
- Assignees: Sick AG
- Applicants: Sick AG
- Inventors: Christian MULTER
- Web page: https://patentgrep.com/patent/US20230419529A1

## CPC

- G01G 19/4144: {for controlling weight of goods in commercial establishments, e.g. supermarket, P.O.S. systems}
- G06T 7/62: of area, perimeter, diameter or volume
- G06T 5/40: using histogram techniques
- G06T 7/11: Region-based segmentation
- G06T 7/50: Depth or shape recovery
- G06V 10/764: using classification, e.g. of video objects
- G06V 20/52: Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69)
- G06V 20/64: Three-dimensional [3D] objects
- G01B 11/24: for measuring contours or curvatures
- G06T 2207/10028: Range image; Depth image; 3D point clouds
- G01G: WEIGHING (sorting by weighing B07C5/16)
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G01B: MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS

## Abstract

The invention relates to a method and apparatus for collecting master data. The device comprises: a scale, a camera for recording an image of the object, an evaluation unit with a volume determination unit for determining the volume of the object, a starter unit which automatically determines a start time for recording the master data relating to the object volume when the object is placed on the scale, a memory for storing the recorded master data, an output unit for outputting the master data, the starter unit being designed to determine the start time from the recorded image of the camera, and the starter unit having an image evaluation unit for this purpose, and the image evaluation unit being designed to carry out an object classification by means of which, in addition to the object, extraneous objects are also recognized as permissible or impermissible extraneous objects.

## Claims

### Claim 1

Method for acquiring master data of an object (12) comprising the steps:
weighing the object on a scale (14),
determining a starting point in time for recording the master data by:
taking a depth value image of the object (12) with a camera (16),
determining contiguous regions in the image that then form first temporary objects,
classifying the first temporary objects into a valid and an invalid class, wherein all first temporary objects that lie within the image without touching the image edge (33) are sorted into the valid class and first temporary objects with edge contact are sorted into the invalid class,
analyzing the first temporary objects in the invalid class by evaluating the depth values and forming new subregions with depth values beyond a depth value threshold,
scanning the subregions for contiguous areas, where these contiguous areas form second temporary objects,
moving the second temporary objects to the valid class if they do not touch the image edge (33),
finding overlaps of the temporary objects in the valid class with the temporary objects remaining in the invalid class, where the overlap must exceed an overlap threshold,
in case of overlap, moving the temporary objects in question from the valid class to the invalid class,
defining the temporary objects in the valid class as the object to be measured,
if no object is in the valid class, output an error signal, otherwise
after the starting point in time determined in this way, starting the volume detection of the object (12),
recording the master data,
saving the recorded master data,
outputting of the master data.

### Claim 2 (depends on 1)

Method according to claim 1, characterized in that depth value histograms are recorded and evaluated for the evaluation of the depth values and formation of new subregions with depth values beyond a depth value threshold.

### Claim 3 (depends on 1)

Method according to claim 1, characterized in that, in the presence of regions with missing depth values, the search for overlaps also comprises the search for overlaps of the temporary objects in the valid class with the regions with missing depth values and, in the case of overlaps, the relevant temporary object is moved from the valid class to the invalid class.

### Claim 4 (depends on 1)

Device for acquiring master data of an object (12) for carrying out a method according to claim 1, comprising
a scale (14),
a camera (16) for capturing an image of the object (12),
an evaluation unit (18) with a volume determination unit (20) for determining the volume of the object (12),
a starter unit (26) which automatically determines a starting point in time for recording the master data concerning the object volume when the object (12) is placed on the scale (14),
a memory (28) for storing the recorded master data,
an output unit (22) for outputting the master data,
characterized in that
the starter unit (26) is designed to determine the starting point in time from the recorded image of the camera (16), and the starter unit (26) has an image evaluation unit (30) for this purpose, and the image evaluation unit (30) is designed to carry out an object classification by means of which, in addition to the object (12), extraneous objects (50) are also classified as permissible or impermissible extraneous objects (50).

### Claim 5 (depends on 4)

Device according to claim 4, characterized in that the starter unit is designed to output the starting point in time as early as possible and already outputs it when a permissible extraneous object is still located in the image area.

### Claim 6 (depends on 4)

Device according to claim 4, characterized in that the camera is adapted to generate an image comprising depth values.

### Claim 7 (depends on 6)

Device according to claim 6, characterized in that the image evaluation unit is designed to perform the object classification taking into account the depth values.

### Claim 8 (depends on 4)

Device according to claim 4, characterized in that the evaluation unit is designed to output a message if no starting point in time could be determined within a waiting time.
