Surround View Back-Projection Removal via Adaptive Processing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Surround view systems for vehicles fail to effectively remove undesired back-projections of moving parts, such as wheels and other components, from the generated images, which can be distracting and disrupt the overall view of the vehicle environment.

Innovation Solution

A surround view system equipped with multiple cameras and a processor that analyzes the position and movement of movable vehicle parts, generates a composite image, and adapts the image processing area to include back-projections, using methods like spatial prediction of color information and motion-compensated texture to remove or modify these projections within the image processing area.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If the image processing area is enlarged to include back-projections of moving parts, then the back-projections can be removed by image processing methods, but the overall image processing area becomes unnecessarily large

Engineering Contradiction:
Improveremoval precision of back-projectionsVSAvoidimage processing area
Core Design Contradiction:
Manufacturing precisionVSArea of stationary object

Solution Approach 1:

The patent applies local quality by modifying the image processing area only in specific local regions where back-projections occur, rather than enlarging the entire processing area. The processor identifies exact locations of back-projections and adjusts the processing area selectively at those positions, maintaining minimal overall processing area while ensuring precise removal of artifacts where needed.

Inventive Principle:
Principle #3Local quality

2Manufacturing precision

If adaptive image processing methods are applied to remove back-projections, then the clarity of the surround view image is improved, but the device complexity increases

Engineering Contradiction:
Improveimage clarityVSAvoidprocessor complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent applies preliminary action by detecting the position and movement of moving parts (such as wheels) before generating the composite surround view image. The processor uses this advance information to predict where back-projections will occur and pre-adjusts the image processing area accordingly, enabling efficient removal of artifacts without requiring complex real-time processing during image generation.

Inventive Principle:
Principle #10Preliminary action

3Productivity

If the image processing area is kept small for efficiency, then processing speed is improved, but back-projections of moving parts cannot be fully removed

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidback-projection removal completeness
Core Design Contradiction:
ProductivityVSManufacturing precision

Solution Approach 1:

The patent applies dynamics by making the image processing area adaptive and dynamic rather than static. The processor continuously adjusts the processing area boundaries based on the real-time position and movement of vehicle parts, expanding the area only where and when back-projections occur and contracting it elsewhere, thus maintaining high processing efficiency while ensuring complete artifact removal.

Inventive Principle:
Principle #15Dynamics

Data Source

PatentEP3610643B1Surround view system for a vehicle
Publication Date: 2021.08.18 CONTI TEMIC MICROELECTRONIC GMBH
  • EP3610643B1 patent drawingFigure 1~2
  • EP3610643B1 patent drawingFigure 3~4

AI summary

The invention relates to a surround view system for a vehicle (1). Said surround view system comprises a processor (10) and a plurality of cameras (2 to 5) that can be arranged on the vehicle (1) in such a way that the plurality of cameras (2 to 5) can record images of an outside environment (6 to 9) of the vehicle (1). The processor (10) is designed to analyse a position (a) and/or a movement of a movable part (14) of the vehicle and to generate a composite image of the outside environment (6 to 9) of the vehicle (1) from images of the outside environment (6 to 9) of the vehicle (1), recorded by the plurality of cameras (2 to 5). Furthermore, the processor (10) is designed to calculate an image processing region within the composite image for an adaptive image processing, to determine, based on the analysed position and/or movement of the movable part (14), that a back-projection of the movable part (14) goes beyond the image processing region, and to modify the image processing region in such a way that the back-projection of the movable part (14) is within the modified image processing region.