Head-Up Display Occlusion Reduction via Warped Zone Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing methods for reducing occlusions by objects in virtual images of head-up displays fail to reliably detect occlusions when the image is pre-distorted, leading to false negatives and increased computational complexity.

Innovation Solution

The method involves warping the image before dividing it into zones, ensuring the center row of each column aligns with the horizontal center line of the warped image, and adapting column heights and thresholds to account for optical distortions, allowing for flexible and efficient occlusion detection even in warped images.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the image is divided into zones after pre-distortion (warping), then the occlusion detection can be performed on the warped image, but the detection reliability deteriorates because the zone alignment no longer corresponds to the horizontal reference line

Engineering Contradiction:
Improveocclusion detection reliabilityVSAvoidimage processing complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies pre-distortion (warping) to the image before dividing it into zones. This preliminary action compensates for optical distortions introduced by the head-up display system, ensuring that the subsequent zone-based occlusion detection remains reliable. The warping is performed in advance so that when zones are divided and occlusion detection is performed, the horizontal reference line alignment is preserved despite the warped geometry.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If zone-specific thresholds are used to improve occlusion detection accuracy, then detection precision improves, but memory capacity requirements increase significantly

Engineering Contradiction:
Improveocclusion detection precisionVSAvoidmemory capacity
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent applies different thresholds to different zones based on their position and characteristics. Instead of using a single uniform threshold for the entire image, zone-specific thresholds are applied to account for variations in occlusion probability across different regions. This local differentiation improves detection precision while managing memory requirements through efficient threshold storage strategies.

Inventive Principle:
Principle #3Local quality

3Reliability

If the image is divided into multiple zones with different heights to account for warping, then occlusion detection accuracy improves, but computational complexity increases

Engineering Contradiction:
Improveocclusion detection accuracyVSAvoidcomputational efficiency
Core Design Contradiction:
ReliabilityVSProductivity

Solution Approach 1:

The patent segments the warped image into multiple zones with different heights, where zones closer to the horizon have smaller heights and zones farther from the horizon have larger heights. This segmentation adapts the zone geometry to match the warped image characteristics, improving occlusion detection accuracy. The segmentation is performed in a way that maintains computational efficiency by leveraging the structured nature of the zone division.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP4531028A1Method for reducing occlusions by objects of a virtual image of a head-up display, computer program code and apparatus therefor
Publication Date: 2025.04.02 CONTINENTAL AUTOMOTIVE TECHNOLOGIES GMBH
  • EP4531028A1 patent drawingFigure 1
  • EP4531028A1 patent drawingFigure 2~3
  • EP4531028A1 patent drawingFigure 4~5

AI summary

The invention is directed to a method for reducing occlusions by objects of a virtual image of a head-up display, a computer program code and an apparatus therefor. The method comprises: - Dividing an image (BD,BA) to be displayed in multiple zones (70,71,72,73), wherein these zones (70,71,72,73) - have rectangular shape, - are arranged in rows (Rn) and columns (700,700-n), and wherein a centre row of said rows (Rn) substantially coincides with a horizontal reference line (642); - Detecting (S4) a fraction (FC) of each zone (70,71,72,73) that is covered by objects (63) of the image (BD,BA) to be displayed; - Warning (S5) by generating and outputting of a warning signal () in case that at least for one of said zones (70,71,72,73) the detected fraction (FC) exceeds a threshold (THR(n)); - preceding the Dividing performing Warping (S1) the image (BD,BA) to be displayed; and the Dividing comprising -Vertical Dividing (S2) the image (BD,BA) to be displayed into columns (700,700-n), and - Adapting (S3) each column (700,700-n) of zones (70,71,72,73) such that the zone (72) of the respective centre row of each column (700,700-n) substantially coincides with a horizontal centre line (647) of the warped image (WB) to be displayed.