18463365. IMAGE PROCESSING SYSTEM AND METHOD FOR RECOGNITION PERFORMANCE ENHANCEMENT OF CIPV simplified abstract (HYUNDAI MOBIS CO., LTD.)
Contents
IMAGE PROCESSING SYSTEM AND METHOD FOR RECOGNITION PERFORMANCE ENHANCEMENT OF CIPV
Organization Name
Inventor(s)
Jae Young Lee of Icheon-si (KR)
IMAGE PROCESSING SYSTEM AND METHOD FOR RECOGNITION PERFORMANCE ENHANCEMENT OF CIPV - A simplified explanation of the abstract
This abstract first appeared for US patent application 18463365 titled 'IMAGE PROCESSING SYSTEM AND METHOD FOR RECOGNITION PERFORMANCE ENHANCEMENT OF CIPV
Simplified Explanation: The technology described in the patent application aims to improve the recognition distance of a vehicle's own lane while reducing the total number of image pixels by adjusting the decimation factor of each region of forward image data. This enhancement focuses on improving the performance of recognizing the Closest In-Path Vehicle (CIPV) region through an object recognition network.
- The technology enhances the recognition distance of a vehicle's own lane.
- It reduces the total number of image pixels by setting appropriate decimation factors for each region of forward image data.
- The focus is on improving the performance of recognizing the Closest In-Path Vehicle (CIPV) region through an object recognition network.
Potential Applications: 1. Autonomous vehicles 2. Advanced driver-assistance systems (ADAS) 3. Traffic monitoring systems
Problems Solved: 1. Limited recognition distance of a vehicle's own lane 2. High computational load due to a large number of image pixels 3. Inefficient object recognition in the Closest In-Path Vehicle (CIPV) region
Benefits: 1. Improved safety on the road 2. Enhanced accuracy in object recognition 3. Reduced computational resources required
Commercial Applications: Enhancing the performance of ADAS systems for various vehicle manufacturers, improving the efficiency of traffic monitoring systems for transportation authorities, and optimizing autonomous vehicles for commercial use.
Questions about the Technology: 1. How does the technology determine the appropriate decimation factor for each region of forward image data? 2. What impact does this technology have on the overall efficiency of object recognition networks in vehicles?
By optimizing the decimation factor of each region of forward image data, the technology described in the patent application aims to enhance the recognition distance of a vehicle's own lane while reducing the total number of image pixels. This innovation has potential applications in autonomous vehicles, ADAS, and traffic monitoring systems, addressing issues such as limited recognition distance, high computational load, and inefficient object recognition. The benefits include improved safety, accuracy, and reduced computational resources, making it valuable for commercial applications in the automotive industry.
Original Abstract Submitted
Provided is technology to enhance a recognition distance of the own lane of a vehicle while reducing a total number of image pixels by appropriately setting a decimation factor of each region of forward image data for recognition performance enhancement of a Closest In-Path Vehicle (CIPV) region by an object recognition network.
