Optical Detection of Empty Beverage Crates Using Illumination Model Correction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing empties return systems face challenges in accurately recognizing and classifying beverage crates with varying designs and labels due to spatial distortion and variable illumination, making it difficult to distinguish between different types of crates and their origin.

Innovation Solution

The system employs a method using at least two cameras to capture geometric parameters and source image data, with a lighting model correction to normalize the image, accounting for geometry and position, and combines multiple exposures to enhance image dynamic and correct for reflections and distortions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If a single camera is used to capture images of beverage crates, then the device complexity is reduced, but spatial distortion and variable illumination cause unreliable object feature recognition

Engineering Contradiction:
Improveobject feature recognition reliabilityVSAvoidcamera system complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent divides the imaging system into multiple independent camera units, each capturing specific geometric or optical information. The first camera captures source images of object features while second cameras capture geometric parameters, allowing independent optimization of each camera's function and reducing the complexity of any single camera system while improving overall recognition reliability through divided functional responsibilities.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an illumination model as an intermediary component that mediates between the captured images and the final object recognition. This illumination model corrects for spatial distortion and variable illumination effects by calculating corrected image data from source images, thereby improving recognition reliability without requiring complex camera systems.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If multiple cameras are used to capture geometric parameters and source images, then classification accuracy is improved, but the device complexity increases

Engineering Contradiction:
Improveclassification accuracyVSAvoidcamera system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the imaging function into specialized cameras: first cameras for capturing source images of object features and second cameras for capturing geometric parameters. This segmentation allows each camera type to be optimized for its specific purpose, improving measurement precision while keeping individual camera systems relatively simple.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a multi-functional system where the combination of first and second cameras serves multiple purposes: capturing geometric data, capturing optical features, correcting for distortion, and enabling accurate classification. This universal approach improves measurement precision through functional integration without proportionally increasing device complexity.

Inventive Principle:
Principle #6Universality (Multi-functionality)

3Manufacturing precision

If image correction using illumination model is applied, then inhomogeneous illumination and spatial distortion are removed, but processing time increases

Engineering Contradiction:
Improveimage data qualityVSAvoidimage processing time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary actions by capturing both source images and geometric parameters simultaneously using multiple cameras. The illumination model is pre-calculated based on the geometric parameters captured in advance, allowing the correction process to proceed efficiently without requiring time-consuming real-time calculations, thus improving image data quality while minimizing processing time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent creates a corrected copy of the source image data by calculating corrected image data from the source images using the illumination model. This copying process preserves the essential information while removing distortion and illumination artifacts, improving manufacturing precision through data transformation rather than physical re-capture, thereby reducing time loss.

Inventive Principle:
Principle #26Copying

Data Source

PatentEP2297674B1Optical detection and classification of empty objects in a system for returning empties
Publication Date: 2015.01.21 WINCOR NIXDORF INT GMBH
  • EP2297674B1 patent drawingFigure 1a
  • EP2297674B1 patent drawingFigure 1b
  • EP2297674B1 patent drawingFigure 1c

AI summary

A method (100) for optically detecting and classifying empty objects (BC) in a system for returning empties (RS) is proposed, wherein the respective empty object (BC) has at least one optical object feature (L) and is introduced into a measuring space (MR) of the system for returning empties (RS). In the method, at least one first camera (LC) is used to record (110) source image data (D-PIC) of the optical object feature (L) and at least one optical sensor system, for example a second camera (C), is used to record (115) geometrical parameters (GP) of the empty object (BC), wherein corrected image data (D-PIC*) are calculated (150) from the source image data (D-PIC) using an illumination model (IMOD), which reproduces the illumination acting on the second camera (LC) in the measuring space (MR) for a model object having the geometrical parameters (GP), in order to identify the optical object feature (L) and to detect and classify that of the respective empty object (BC).