Phase-Difference Distance Sensing for Non-Planar Object Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing distance measurement systems for vehicles face challenges in accurately measuring distances to objects due to large disparities between camera fields and non-planar object characteristics, leading to incorrect detection of corresponding points and erroneous distance measurements.
Innovation Solution
A distance measuring apparatus utilizing monocular cameras with phase-difference detecting pixels and a processor that calculates distances using phase-difference image data, combined with reliability assessment to ensure accurate measurements, and integrates monocular and stereoscopic measurements to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If multiple cameras are installed around a vehicle to expand imaging coverage, then the coverage area is improved, but the complexity of the system increases and regions with non-overlapping fields remain
Solution Approach 1:
The imaging space is segmented into overlapping regions and non-overlapping regions. Different measurement methods (stereoscopic vs. monocular) are applied to different segments, allowing the system to handle complex multi-camera configurations while maintaining measurement accuracy in each region.
Solution Approach 2:
The distance measuring apparatus is designed to perform multiple functions: stereoscopic distance measurement in overlapping regions and monocular distance measurement in non-overlapping regions. This multi-functional approach allows a single system to handle diverse imaging scenarios without requiring separate systems for each region type.
2Measurement precision
If stereoscopic distance measurement is used in overlapping regions, then measurement accuracy is improved, but incorrect detection occurs for non-planar objects due to large disparities
Solution Approach 1:
The system dynamically selects the measurement method based on object characteristics and region type. For non-planar objects in overlapping regions, it switches from stereoscopic to monocular measurement, adapting the measurement approach to the specific conditions to maintain both accuracy and reliability.
Solution Approach 2:
The system changes the measurement parameters by switching between stereoscopic and monocular modes. This parameter change allows the system to handle different object types (planar vs. non-planar) appropriately, avoiding the large disparity issues that plague stereoscopic measurement of non-planar objects.
3Reliability
If monocular distance measurement is used for non-planar objects, then detection reliability is improved, but measurement accuracy decreases due to lack of depth information
Solution Approach 1:
The system merges stereoscopic and monocular distance measurement results through integration processing. By combining the depth information from stereoscopic measurement with the reliable detection from monocular measurement, the system achieves both accuracy and reliability for non-planar objects.
Solution Approach 2:
The integration processing unit acts as an intermediary that combines results from different measurement methods. It reconciles the strengths and weaknesses of stereoscopic and monocular measurement to produce a final accurate distance measurement for non-planar objects.
4Area of stationary object
If the vehicle body blocks part of the camera field, then the imaging range is reduced, but additional cameras increase system complexity
Solution Approach 1:
Instead of trying to image blocked regions directly, the system inverts the approach by using the visible portions of objects in blocked regions and tracing their positions back to the obscured areas. This allows the system to infer distance information for regions partially blocked by the vehicle body without adding more cameras.
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 accurate distance measurements by correcting light quantity balance, calculating image shifts, and assessing reliability, enabling precise vehicle control and collision avoidance operations.
Implementation Method 1
an image sensor (101) that generates first phase-difference image data and second phase-difference image data by imaging, with phase differences, an object with two different optical paths through the optical system (120)
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
A distance measuring apparatus (2) includes a distance measuring unit for an object based on a phase difference obtained by receiving light from the object using two light receivers (161, 162, 101), the distance measuring unit including a first distance measuring unit (11) with a first base length (w), and a second distance measuring unit (11, 16) with a second base length which is longer than the first base length, an acquiring unit (24, 25) configured to acquire a characteristic of the object, and a generator (102) configured to generate distance information on the object. The generator is configured to generate the distance information using at least one of distance measurement results obtained respectively by the first distance measuring unit and the second distance measuring unit, in accordance with the characteristic.