Parking Assistance Device Using Stereo Image Processing for Obstacle Position Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing parking assistance systems face challenges in accurately determining the target parking position, leading to delayed recognition and increased error due to continuous route changes after entering the parking space and issues with image capture rates and vehicle speed.
Innovation Solution
A parking assistance device that includes a parking space candidate detection unit, an image selection unit, a stereo image processing unit, and a parking space correction unit, which estimates obstacle positions, selects and processes images to calculate three-dimensional point groups, and corrects the parking space candidate position based on these calculations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If continuous route determination is performed during parking assistance, then accuracy of route is improved, but time required for parking increases due to delayed recognition of correct target parking position
Solution Approach 1:
The system performs preliminary determination of the target parking position using ultrasonic sonar to detect the inner side surface of obstacles before parking assistance begins. This preliminary action establishes an initial accurate target position, preventing the need for continuous route adjustments during parking and thereby reducing parking time while maintaining accuracy.
2Measurement precision
If image sampling rate is increased to capture inner side surface, then position determination accuracy is improved, but system complexity and processing load increase
Solution Approach 1:
The system uses ultrasonic sonar as an intermediary device to detect the inner side surface of obstacles and determine obstacle positions. This intermediary approach provides accurate position data without requiring complex high-rate image processing, thereby achieving position determination accuracy while avoiding excessive system complexity.
Solution Approach 2:
The system replaces continuous high-rate image processing with ultrasonic sonar detection for obstacle position measurement. This substitution uses acoustic waves instead of optical imaging, achieving accurate position determination with simpler, less computationally intensive technology.
3Measurement precision
If ultrasonic sonar is used to detect inner side surface, then target parking position accuracy is improved, but reliability decreases when no central image is available due to sampling rate or vehicle speed
Solution Approach 1:
The system continuously monitors the availability of images containing the inner side surface during parking assistance. When such images become available, the system updates the target parking position using stereo image processing to correct and refine the position. This feedback mechanism ensures reliability by adaptively using the most accurate available data throughout the parking process.
Solution Approach 2:
The system dynamically adjusts its position determination strategy based on real-time conditions. It transitions from relying on preliminary ultrasonic sonar data to using stereo image processing when appropriate images become available during parking, adapting to changes in vehicle speed, sampling rate, and image capture conditions to maintain both accuracy and reliability.
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
Enables quick and accurate determination of the parking space candidate position, reducing the time taken for parking and improving accuracy by performing stereo image processing on captured images of the inner side surfaces of obstacles.
Implementation Method 1
an image selection unit that selects two or more captured images in which an inner side surface of an adjacent obstacle adjacent to the parking space candidate on the side of the parking space candidate is captured among images captured by an imaging unit provided on the host vehicle
Implementation Method 2
a stereo image processing unit that performs stereo image processing on the selected two or more captured images and calculates a three-dimensional point group of the obstacle
Data Source
AI summary
A parking assistance device includes: a parking space candidate detection unit estimating positions of obstacles based on data of distance to the obstacles at both sides of a host vehicle, and detecting a parking space candidate at the sides of the vehicle; an image selection unit selecting two or more captured images in which an inner side surface of an obstacle adjacent to the parking space candidate is captured; a stereo image processing unit performing stereo image processing on the two or more captured images and calculating a three-dimensional point group of the obstacle; a position determination unit determining a position of the obstacle on the parking space candidate side based on the calculated three-dimensional point group; and a parking space correction unit correcting a position of the parking space candidate based on the determined position of the obstacle on the parking space candidate side.