Object Assignment in Overlapping Image Sensor Sub-Regions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing monitoring systems face inaccuracies and incorrect object assignments due to large subdivided regions and the need for camera synchronization, which increases costs and complexity.
Innovation Solution
A method and device that use image sensors to assign objects across overlapping sub-regions by rating their correspondence, allowing for improved determination of total objects in a monitoring region, with a calculation unit integrated into or separate from the sensors, and enabling the use of various types of sensors and objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If objects are assigned by means of assignment of overlapping subdivided regions, then object assignment can be performed, but inaccuracies occur and incorrect assignments happen
Solution Approach 1:
The patent changes the parameter of object representation from simple region-based assignment to a comprehensive parameter set including position coordinates, size dimensions, shape characteristics, and color features. This multi-parameter approach enables more accurate matching and assignment of objects across overlapping camera views, resolving the contradiction between assignment capability and assignment accuracy.
2Ease of operation
If video cameras are synchronized to combine image sequences, then a common superordinate image can be formed, but installation costs increase
Solution Approach 1:
The patent extracts the essential information needed for object tracking (position, size, shape, color parameters) from the image sequences without requiring full video synchronization. By taking out only the necessary object parameters rather than synchronizing entire video streams, the system achieves image combination capability while avoiding the complexity and cost of camera synchronization hardware and software.
3Speed
If objects are modeled as pixel accumulations and tracked, then movement can be monitored, but the system becomes more complex
Solution Approach 1:
The patent changes the modeling approach from pixel accumulation to parameter-based representation, where objects are characterized by key parameters (position coordinates, size, shape, color) rather than individual pixels. This parameter change enables efficient tracking of object movement while significantly reducing system complexity, as the tracker operates on compact parameter sets rather than large pixel data sets.
Data Source
AI summary
A method and a device for monitoring a monitoring region with at least two image sensors. A sub-region of the monitoring region is monitored by each of the image sensors, wherein each image sensor detects objects to be monitored that are located within the sub-region monitored by said image sensor, and each image sensor outputs data relating to the detected objects and are disposed and oriented in such a way that the monitored sub-regions overlap and that each object to be monitored that is located in the monitoring region is always detected by at least one image sensor.