Surround-View Image Projection for Low-Burden Obstacle Rendering

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing systems for vehicles face increased processing burdens when generating composite images with multiple obstacles, leading to unnaturalness and inefficiency in viewpoint conversion.

Innovation Solution

An image processing device that determines combination patterns based on the presence of vehicles in parking spaces around the vehicle and selects appropriate projection surfaces to generate composite images, reducing processing burdens and unnaturalness by optimizing image projection and rendering.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If standard mapping table information and compression mapping table information are used in combination based on obstacle distance, then the unnaturalness of the composite image is reduced, but the processing burden increases

Engineering Contradiction:
Improveimage qualityVSAvoidprocessing burden
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent segments the image processing into two distinct modes: standard mapping for regions without obstacles and compression mapping for regions with obstacles. The viewpoint conversion processor selectively applies different mapping table information to different image regions based on obstacle detection results, thereby reducing unnaturalness while managing processing complexity through structured segmentation of the processing workflow

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system dynamically switches between standard and compression mapping table information based on the detected obstacle distance. When an obstacle is detected within a predetermined distance, the processor dynamically adjusts the mapping approach for the affected image region, enabling adaptive image quality optimization without permanently increasing system complexity

Inventive Principle:
Principle #15Dynamics

2Manufacturing precision

If multiple mapping table information are used for different obstacle distances, then the composite image quality improves, but the processing time increases

Engineering Contradiction:
Improvecomposite image qualityVSAvoidviewpoint conversion time
Core Design Contradiction:
Manufacturing precisionVSLoss of time

Solution Approach 1:

Multiple mapping table information (standard and compression mapping tables) are prepared in advance and stored in the storage unit. This preliminary preparation allows the viewpoint conversion processor to quickly switch between pre-computed mapping tables based on obstacle detection, avoiding the need to compute new mapping relationships in real-time and thereby reducing processing time while maintaining image quality

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system changes the mapping parameters by selecting different pre-prepared mapping table information based on the detected obstacle distance. Instead of continuously adjusting mapping parameters, the system switches between discrete parameter sets (standard vs. compression mapping tables), which simplifies the real-time processing while still adapting to different scene conditions

Inventive Principle:
Principle #35Parameter changes

Data Source

PatentEP3761262B1Image processing device and image processing method
Publication Date: 2024.06.12 DENSO CORP
  • EP3761262B1 patent drawingFigure 1
  • EP3761262B1 patent drawingFigure 2
  • EP3761262B1 patent drawingFigure 3

AI summary

An image processing device used on a mobile body has: a determiner configured to determine one out of a plurality of combination patterns as to the presence or absence of an object in a plurality of predetermined regions around the mobile body; and a generator configured to generate a composite image by projecting onto a virtual projection surface a plurality of shot images acquired by a plurality of cameras taking images of the surroundings of the mobile body. The generator is configured to select, out of a plurality of projection surfaces prepared beforehand, the projection surface used in generating the composite image in accordance with the combination pattern determined by the determiner.