Depth Camera Driving Assistance System for Low Light Object Detection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current vehicle driving assistance systems rely on radar or monocular cameras, which have limitations such as inability to distinguish obstacle types and inaccurate distance estimation, especially in low light conditions.

Innovation Solution

A driving assistance system using a depth camera based on active light and an imaging camera to generate driving assistance information by detecting object positions in a scene image, enabling accurate recognition of objects and lane information even at night or in low light.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If radar is used to detect obstacles, then distance detection is achieved, but obstacle category distinction and lane status determination are not possible

Engineering Contradiction:
Improvedistance detection accuracyVSAvoidobstacle category information
Core Design Contradiction:
Measurement precisionVSLoss of information

Solution Approach 1:

The patent combines radar and camera systems into a unified detection framework. The radar provides distance measurement while the camera provides obstacle category identification and lane status determination. By merging these two systems, the patent resolves the contradiction by allowing each system to compensate for the other's limitations through information fusion.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces an information fusion module as an intermediary that processes data from both radar and camera systems. This intermediary component integrates the distance information from radar with the categorical information from camera, enabling the system to achieve both precise distance measurement and obstacle identification simultaneously.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Loss of information

If monocular camera is used to recognize objects, then object classification is achieved, but distance estimation accuracy is insufficient

Engineering Contradiction:
Improveobject classification informationVSAvoiddistance estimation accuracy
Core Design Contradiction:
Loss of informationVSMeasurement precision

Solution Approach 1:

The patent merges monocular camera and stereo camera systems to overcome the limitations of single-system operation. The monocular camera provides object classification while the stereo camera provides accurate distance measurement. By combining these systems, the patent enables the system to simultaneously achieve both object recognition and precise distance estimation.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces a data fusion module as an intermediary that integrates information from both monocular and stereo cameras. This intermediary processes the complementary information from both camera systems, enabling accurate object classification and distance measurement through combined data analysis.

Inventive Principle:
Principle #24Intermediary (Mediator)

3Loss of information

If passive light-based cameras are used for detection, then object recognition is achieved, but detection capability deteriorates in low light or night conditions

Engineering Contradiction:
Improveobject recognition informationVSAvoiddetection reliability in low light
Core Design Contradiction:
Loss of informationVSReliability

Solution Approach 1:

The patent combines active light-based radar systems with passive light-based camera systems. The radar system using active light can detect obstacles in low light or night conditions, while the camera system provides detailed object recognition. By merging these systems, the patent maintains reliable detection across various lighting conditions while preserving object recognition capabilities.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent introduces a multi-sensor fusion module as an intermediary that integrates data from both active light-based radar and passive light-based cameras. This intermediary component processes the complementary strengths of both systems, enabling reliable detection in low light conditions while maintaining accurate object recognition through combined sensor data.

Inventive Principle:
Principle #24Intermediary (Mediator)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

The system provides more accurate driving assistance information compared to prior art, ensuring reliable object detection and distance estimation, enhancing traffic safety and efficiency.

Implementation Method 1

a depth camera based on active light

Methodology Applied
Scientific EffectActive light emission and reflection: Light

Implementation Method 2

an imaging camera based on passive light

Methodology Applied
Scientific EffectPassive light detection: Light

Data Source

PatentUS10255510B2Driving assistance information generating method and device, and driving assistance system
Publication Date: 2019.04.09 BEIJING KUANGSHI TECHNOLOGY CO LTD
  • US10255510B2 patent drawing
  • US10255510B2 patent drawing
  • US10255510B2 patent drawing

AI summary

Provided are a driving assistance information generating method and device, and a driving assistance system, which relate to a field of vehicle driving assistance technique. The driving assistance information generating method comprises: obtaining a depth image acquired by a depth camera and a scene image acquired by an imaging camera, the depth camera and the imaging camera being mounted on a vehicle in a manner of being registered and matched with each other, and a coverage of the depth camera and a coverage of the imaging camera being at least partially overlapped; detecting positions of respective objects appearing in the scene image by using the depth image and the scene image; and generating driving assistance information of the vehicle based on positions of the respective objects.