Forward Vehicle Detection Using ROI Tracking for Faster ADAS
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle detection systems in advanced driver assistance systems (ADAS) face challenges in accurately and stably detecting forward vehicles from captured driving images, which is crucial for improving the accuracy of collision warning systems.
Innovation Solution
A method that involves continuously capturing front images, setting a search area for vehicle detection based on the location or area detected in previous images, using a machine learning model for vehicle detection, and tracking vehicles using feature points extracted from previous images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vehicle detection is performed using machine learning model on the entire captured image, then detection accuracy is improved, but processing speed deteriorates
Solution Approach 1:
The captured image is divided into multiple regions of interest (ROIs) based on detected feature points from previous frames. The machine learning model is applied only to these segmented ROIs rather than the entire image, reducing computational load while maintaining detection accuracy for vehicles in critical areas.
Solution Approach 2:
Different processing quality is applied to different regions of the image. Regions containing feature points or previously detected vehicles receive full machine learning model processing for high accuracy, while other regions use simplified detection methods, optimizing the balance between accuracy and processing speed.
2Reliability
If search area is enlarged to cover all possible vehicle locations, then detection reliability is improved, but processing time increases
Solution Approach 1:
Feature points are extracted and tracked from previous frames before performing vehicle detection in the current frame. This preliminary action identifies likely vehicle locations in advance, allowing the search area to be focused on these pre-identified regions rather than scanning the entire image, thus maintaining reliability while reducing processing time.
Solution Approach 2:
The search area dynamically adjusts based on the movement and position of feature points from previous frames. The search region expands or contracts according to the tracked feature points, ensuring comprehensive coverage of potential vehicle locations while minimizing unnecessary processing of empty areas.
3Stability of the object's composition
If tracking is performed continuously using feature points, then vehicle location stability is improved, but computational complexity increases
Solution Approach 1:
Feature point extraction and tracking are performed continuously across multiple frames to maintain stable vehicle location information. This continuous tracking ensures that vehicle positions remain stable and consistent over time, providing reliable data for detection and warning systems.
Solution Approach 2:
Instead of performing complex detection on every frame, the system creates simplified copies or representations of vehicle locations using tracked feature points. These feature point copies provide stable location information with lower computational requirements than full image processing.
Data Source
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.