Multi-Camera Vision System Object Detection with Segmented Zones
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vision systems for monitoring vehicle surroundings, especially in the automotive market, face challenges with high costs, power consumption, and complexity in rapidly processing large amounts of data for real-time responses, making them unsuitable for widespread deployment in driver-operated vehicles for enhanced traffic safety.
Innovation Solution
The method involves acquiring scene images from multiple zones around a moving vehicle using image acquisition devices with varying field of view angles and depth of field settings, applying template matching criteria and scan windows of different sizes to identify objects, and classifying them based on size and position relative to the frame of image data, allowing for efficient detection and classification of objects like pedestrians and vehicles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If LiDAR systems are used to create comprehensive dynamic maps of vehicle surroundings, then measurement precision and reliability are improved, but device complexity and cost increase significantly
Solution Approach 1:
The monitoring region is divided into multiple zones (first zone, second zone, etc.) with different depth of field settings. Each zone is monitored by camera devices with specific focal length ranges, allowing the system to process data more efficiently by segmenting the visual field into manageable regions rather than attempting to process the entire scene uniformly.
Solution Approach 2:
Different camera devices are assigned different focal length ranges and depth of field settings optimized for specific zones. The first camera device monitors the first zone with a first focal length range, while the second camera device monitors the second zone with a second focal length range, allowing each device to operate at optimal quality for its designated region.
2Measurement precision
If multiple camera devices with different depth of field settings are used to monitor different zones, then measurement precision is improved, but device complexity increases
Solution Approach 1:
Multiple camera devices perform the same basic function of monitoring and capturing images, but each is optimized for different focal length ranges and zones. This multi-functional approach allows the system to maintain precision across varying distances while using standardized camera components rather than requiring entirely different sensor types.
Solution Approach 2:
The system varies the focal length parameter and depth of field settings across different camera devices to optimize performance for different zones. By changing these optical parameters rather than using a single fixed configuration, the system achieves high measurement precision for objects at various distances without requiring complex adaptive mechanisms in each device.
3Productivity
If comprehensive monitoring of all zones is performed with high processing power, then productivity is improved, but use of energy increases
Solution Approach 1:
The system segments the monitoring task by assigning different camera devices to different zones with specific focal length ranges. This segmentation allows the central processing unit to receive pre-filtered data from each zone, reducing the overall computational burden and energy required for processing while maintaining real-time monitoring capability across all zones.
Solution Approach 2:
Camera devices perform preliminary action by capturing and pre-processing images within their designated zones and focal length ranges before transmitting data to the central processing unit. This preliminary processing at the source reduces the amount of raw data that requires high-power processing later, enabling real-time responses with lower overall energy consumption.
Data Source
AI summary
Methods for detecting, identifying and displaying object information with a multi-camera vision system. In one embodiment presence of an object is detected in a region extending above a surface and positioned beyond a border, the region including at least first and second zones. A frame of image data is acquired which spans a field of view angle corresponding to a portion of a region from a position along the border. Object detection criteria are applied to analyze and classify an image of an object among a set of object types. A determination is made whether the object is present in the first zone based on application of a first series of scan windows varying in size over a first range of window sizes relative to the size of the image frame, this only permitting detection of images limited in size to a first range of image sizes.