Vehicle Camera Depth-of-Field Fusion for Object Distance Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing optical systems on vehicles struggle to accurately determine the distance of objects in a wide field of view, as they often lose distance information for multiple objects within the field of view, which is crucial for effective vehicle response.

Innovation Solution

A vehicle system equipped with multiple cameras, each having different depth of fields, captures and processes image data to detect objects in distinct in-focus regions, allowing for precise distance determination and object characterization using edge detection and machine learning algorithms.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a two-dimensional sensor array is used to capture image data, then height and length measures of objects can be resolved, but distance measures of objects cannot be accurately determined

Engineering Contradiction:
Improvedistance measurementVSAvoiddistance information
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent transitions from two-dimensional image capture to three-dimensional spatial understanding by introducing depth as an additional dimension. Multiple cameras positioned at different locations capture images from different perspectives, enabling the system to reconstruct depth information and calculate actual distances to objects in the environment.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The patent introduces an intermediary processing system that receives images from multiple cameras, performs image registration and matching, and calculates distance information. This intermediary computational layer transforms 2D image data into 3D spatial understanding, bridging the gap between simple image capture and accurate distance measurement.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Measurement precision

If multiple cameras with different depth of fields are used to capture image data, then object detection accuracy improves, but system complexity increases

Engineering Contradiction:
Improveobject detection accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent divides the imaging task into segments by using multiple cameras, each with a specific depth of field optimized for detecting objects at particular distances. This segmentation allows each camera to excel at detecting objects within its optimal range, improving overall detection accuracy across various distances while maintaining manageable individual camera complexity.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent creates a universal detection system where multiple cameras with different depth of fields work together to provide comprehensive object detection across all distance ranges. The system integrates data from all cameras to achieve multi-functional capability, detecting both near and far objects with high accuracy through a unified processing framework.

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

Data Source

PatentEP4700720A1System and method of off-board object detection by a vehicle
Publication Date: 2026.02.25 TRANSPORTATION IP HOLDINGS LLC
  • EP4700720A1 patent drawingFigure 1
  • EP4700720A1 patent drawingFigure 2A
  • EP4700720A1 patent drawingFigure 2B

AI summary

A control circuit that receives first image data from a first camera on a vehicle. The circuit, using the first image data, determines a first in-focus region defined by a first depth of field (DoF) within a first image frame. The control circuit detects an object in the first in-focus region. The control circuit receives second image data from a second camera on the vehicle. Using the second image data, the control circuit detects the object in a second in-focus region, defined by a second DoF, within a second image frame based on the first image data. The second DoF may be greater than the first DoF. The control circuit determines a characteristic of the object from the second image data, and may operate the vehicle using the vehicular operational data.