ToF Sensor Foreground Extraction via Depth Amplitude Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing depth camera systems face challenges in accurately extracting foreground object data due to noise in depth data, which can be misinterpreted as foreground objects, leading to incorrect identification.
Innovation Solution
An image processing device and method that generate foreground object data by performing dual detection operations using both depth and amplitude data, with reference background data, and include noise removal processes to enhance accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If only depth data is used for foreground object detection, then the detection process is simple and fast, but noise in depth data is misinterpreted as foreground objects leading to incorrect identification
Solution Approach 1:
The patent combines depth data and amplitude data from the same ToF sensor to perform foreground object detection. By merging multiple data types, the system achieves more reliable identification while using a single sensor device, thus improving accuracy without proportionally increasing device complexity.
Solution Approach 2:
The patent introduces reference background data as an intermediary element to compare against captured depth and amplitude data. This reference data serves as a baseline to distinguish actual foreground objects from noise, improving detection reliability through comparative analysis.
2Measurement precision
If dual detection operations using both depth and amplitude data are performed, then foreground object detection accuracy is improved, but processing time and computational complexity increase
Solution Approach 1:
The patent performs preliminary generation of reference background data for both depth and amplitude in advance, before actual foreground object detection is needed. This pre-processing step allows the detection phase to proceed more quickly by comparing against pre-established references rather than computing baselines in real-time.
Solution Approach 2:
The detection process is segmented into distinct operations: depth data detection and amplitude data detection. These segmented detection operations can be performed independently and their results combined, allowing for more efficient processing compared to a monolithic detection approach.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
The solution effectively reduces noise influence, allowing for more accurate extraction of foreground object data by combining depth and amplitude data analysis, improving the reliability of foreground object identification.
Implementation Method 1
A typical method for obtaining the depth data is a Time-of-Flight (ToF) method. A ToF system may include a ToF system that issues a modulated signal to an object and receives a signal reflected from the object. The ToF system measures time spent while the reflected signal is being returned
Data Source
AI summary
An image processing device capable of generating foreground object data by using a captured image includes a depth data generator configured to generate depth data of the captured image; an amplitude data generator configured to generate amplitude data of the captured image; a foreground object detector configured to perform a first detection operation for detecting the foreground object based on the generated depth data and first reference background data, and to perform a second detection operation for detecting the foreground object based on the generated amplitude data and second reference background data; and a foreground object data generator configured to generate the foreground object data based on a result of the first detection operation and a result of the second detection operation.