Image Processing Filter Selection for Rapid Vehicle Speed Convergence

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing image processing systems face significant errors in distance and speed calculations immediately after object detection, particularly for distant or oncoming vehicles, due to unknown initial conditions, leading to delayed filter convergence and inaccurate vehicle control.

Innovation Solution

An image processing device employing a parallel filter process with multiple filters, each with different initial speed values, to quickly converge on an accurate filter result by selecting the filter with the smallest accumulated observation residual value.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a single filter process is used for distance and speed measurement, then the device complexity is low, but the measurement precision is poor immediately after object detection

Engineering Contradiction:
Improvedistance and speed measurement precisionVSAvoidfilter processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The filter processing is segmented into multiple parallel filter processes (first filter process and second filter process), each handling different aspects of the measurement. This segmentation allows the system to improve measurement precision by selecting appropriate filter results based on observation residuals, while managing complexity through structured division of filtering tasks.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes filter parameters by performing multiple filter processes with different initial values and configurations. By calculating observation residuals for each filter process and selecting the one with the smallest residual, the system adapts parameter selection dynamically, improving measurement precision without permanently increasing system complexity.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If the filter process is started immediately with arbitrary initial values, then the ease of operation is high, but the reliability is poor due to large errors in initial measurements

Engineering Contradiction:
Improvemeasurement reliabilityVSAvoidinitial value setting complexity
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The system performs preliminary actions by executing multiple filter processes in parallel from the start, each with different initial values. This preliminary parallel processing ensures that reliable measurements are obtained quickly without requiring complex preliminary analysis or manual intervention for initial value selection.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements feedback by calculating observation residuals for each filter process and using this feedback to select the most reliable filter result. This feedback mechanism automatically improves reliability by selecting the filter with the smallest residual, eliminating the need for manual initial value tuning while ensuring measurement accuracy.

Inventive Principle:
Principle #23Feedback

3Measurement precision

If multiple filter processes are performed in parallel, then the measurement precision improves quickly, but the use of energy increases

Engineering Contradiction:
Improvespeed and distance calculation accuracyVSAvoidprocessing energy consumption
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

Multiple filter processes are performed in parallel as a preliminary action immediately after object detection. This approach achieves quick measurement precision convergence without requiring prolonged single-filter processing, thereby reducing total energy consumption despite the parallel execution overhead.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system changes processing parameters by dynamically selecting which filter process results to use based on observation residuals. This parameter selection strategy avoids the need to continuously run all filter processes indefinitely, reducing energy consumption while maintaining high measurement precision through selective result usage.

Inventive Principle:
Principle #35Parameter changes

4Speed

If a filter process is performed with fixed initial values, then the ease of operation is high, but the speed of filter error convergence is slow for distant or oncoming vehicles

Engineering Contradiction:
Improvefilter error convergence speedVSAvoidfilter processing structure
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

The filter processing is segmented into multiple parallel processes with different initial values, allowing the system to achieve faster error convergence by selecting the appropriate segmented filter result based on observation residuals, rather than relying on a single slow-converging filter.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system changes filter parameters by using multiple initial values in parallel filter processes and dynamically selecting the optimal one based on observation residuals. This parameter diversification accelerates error convergence speed without requiring complex adaptive parameter adjustment mechanisms.

Inventive Principle:
Principle #35Parameter changes

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

Reduces speed errors immediately after object detection, enabling precise vehicle control and enhancing the performance of advanced driver-assistance systems like ACC and AEB.

Implementation Method 1

a filter processing unit that performs a filter process on the distance calculated by the distance calculation unit and the speed calculated by the speed calculation unit. The filter processing unit performs a plurality of filter processes for a predetermined period of time and then selects one of the plurality of filter processes.

Methodology Applied
Scientific EffectFilter process:

Data Source

PatentEP4600605A1Image processing device and image processing method
Publication Date: 2025.08.13 ASTEMO LTD
  • EP4600605A1 patent drawingFigure 1
  • EP4600605A1 patent drawingFigure 2
  • EP4600605A1 patent drawingFigure 3

AI summary

An image processing device includes: a distance calculation unit that calculates a distance to a detection target searched from an acquired image; a speed calculation unit that calculates a speed of the detection target; and a filter processing unit that performs a filter process on the calculated distance and speed. Here, the filter processing unit performs a plurality of filter processes for a predetermined period of time and then selects one of the plurality of filter processes.