Forward Vehicle Detection Using Adaptive Search and Tracking

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing vehicle detection systems face challenges in processing speed and stability, particularly when detecting forward vehicles in augmented reality navigation, and in accurately tracking vehicles using machine learning.

Innovation Solution

The system employs an adaptive search area based on the location and size of the detected vehicle, using machine learning models like the Haar algorithm to enhance processing speed and accuracy, and combines this with feature point extraction and optical flow for continuous vehicle tracking.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the entire image is used as a vehicle detection area, then the detection coverage is comprehensive, but the processing speed decreases

Engineering Contradiction:
Improvedetection coverageVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent divides the image processing task into two stages: first, a rough detection phase using a limited search area based on previous vehicle location; second, a precise detection phase using the Haar algorithm on the detected vehicle region. This segmentation allows comprehensive detection coverage while maintaining high processing speed by avoiding full-image processing.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The system performs preliminary action by using the location and size of previously detected vehicles to define a search area for the current frame. This preliminary positioning enables the system to focus subsequent detailed detection only on relevant regions, significantly reducing processing time while maintaining detection accuracy.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If machine learning is applied to vehicle detection, then the detection accuracy is improved, but the processing time increases

Engineering Contradiction:
Improvedetection accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent applies the Haar algorithm (machine learning) only to the specific search area where vehicles are likely to be detected, rather than processing the entire image. This local application of machine learning maintains high detection accuracy while significantly reducing processing time compared to full-image analysis.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system performs partial action by applying machine learning detection only to the predefined search area based on previous vehicle locations, rather than exhaustively processing the entire image. This partial processing achieves sufficient detection accuracy for the navigation purpose while minimizing time loss.

Inventive Principle:
Principle #16Partial or excessive action

3Stability of the object's composition

If vehicle tracking is implemented for continuous display, then the navigation stability is improved, but the system complexity increases

Engineering Contradiction:
Improvenavigation stabilityVSAvoidsystem complexity
Core Design Contradiction:
Stability of the object's compositionVSDevice complexity

Solution Approach 1:

The system implements feedback by continuously tracking detected vehicles across multiple frames and using this tracking information to maintain stable display in augmented reality navigation. The feedback loop ensures that vehicles remain continuously displayed even during transient detection failures, improving navigation stability without requiring complex redundant systems.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3663978B1Method and device for detecting a vehicle
Publication Date: 2025.03.12 THINKWARE
  • EP3663978B1 patent drawingFigure 1~2
  • EP3663978B1 patent drawingFigure 3~4A
  • EP3663978B1 patent drawingFigure 4B

AI summary

There is provided a method for detecting a vehicle including receiving continuously captured front images, setting a search area of the vehicle in a target image based on a location of the vehicle or a vehicle area detected from a previous image among the front images, detecting the vehicle in the search area according to a machine learning model, and tracking the vehicle in the target image by using feature points of the vehicle extracted from the previous image according to a vehicle detection result based on the machine learning model. Since the entire image is not used as a vehicle detection area, a processing speed may be increased, and a forward vehicle tracked in an augmented reality navigation may be continuously displayed without interruption, thereby providing a stable service to the user.