Vehicular Blind Spot Detection Using Attention Area Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Large-size vehicles have significant blind spots due to their height and length, leading to accidents, as existing mirror systems fail to provide comprehensive and timely warnings, and previous camera-based detection systems face issues with image recognition and processor overload.
Innovation Solution
A vehicular collision warning system utilizing image capture devices, processing units, and alert devices to detect blind spots, calculate image complexity, and send graded alerts of images, light, or sound based on the danger level, integrating with speed sensors and brake systems to optimize warning timing and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the picture is divided into smaller blocks to improve recognizability, then detection accuracy is improved, but processor load increases causing delayed warnings
Solution Approach 1:
The image is divided into multiple blocks for analysis, but the patent applies selective segmentation by focusing processing on specific attention areas rather than uniformly processing the entire image. This allows sufficient block-level analysis for accuracy while reducing overall computational load by concentrating resources on critical regions where barriers are most likely to appear.
Solution Approach 2:
The patent implements different processing qualities for different regions of the image by defining attention areas with higher processing priority. Critical regions near the vehicle receive more intensive analysis with smaller block sizes, while less critical distant regions use coarser processing, thereby maintaining detection accuracy where needed while reducing overall processor burden.
2Area of stationary object
If multiple view mirrors are installed to reduce blind spots, then coverage area is improved, but driver's ability to monitor all mirrors simultaneously deteriorates
Solution Approach 1:
The patent replaces the mechanical mirror system with an electronic image capture and processing system. Cameras mounted on the vehicle capture images from blind spot areas, and the processing unit analyzes these images to detect barriers. This substitution eliminates the need for the driver to physically monitor multiple mirrors while maintaining comprehensive blind spot coverage through automated electronic detection.
Solution Approach 2:
The patent introduces an intermediary processing unit that acts as a mediator between the image capture devices and the driver. This processing unit automatically analyzes captured images, identifies barriers in blind spots, and generates warning signals, thereby eliminating the need for direct driver monitoring of multiple mirror surfaces while maintaining situational awareness.
3Difficulty of detecting and measuring
If view mirrors are used to detect blind spots, then detection capability is improved, but active warning capability deteriorates
Solution Approach 1:
The patent implements a feedback mechanism where the processing unit continuously monitors images from capture devices, analyzes detected objects, and provides real-time warning signals to the driver through alert devices. This closed-loop feedback system automatically transforms passive mirror detection into an active warning system that notifies the driver of potential hazards without requiring continuous manual monitoring.
Solution Approach 2:
The processing unit serves as an intermediary that bridges the gap between passive image capture and active driver warning. It automatically processes captured images, identifies barriers, determines their movement and position, and triggers appropriate warning signals, thereby adding active warning capability that mirrors alone cannot provide.
Data Source
AI summary
A vehicular collision warning system comprises at least one image capture device capturing images of blind spots, a processing unit receiving the images of blind spots from the image capture device, and at least one alert device. The processing unit identifies at least one barrier from the images and calculates the speed and transverse displacement of the barrier according to the positional change with respect to time. The processing unit presets at least one attention area on the image, estimates whether the barrier will appear in the attention area, sends out an alert-triggering signal to trigger the alert devices to generate sound, light, or pictures to warn the driver early. Cooperating with the speed sensor, brake system and turn light system, the vehicular collision warning system can trigger different alert signals according to the grading of danger to actively warn the driver early.