Sensor Data Filtering for Structural Feature Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Enforcement, security, and military forces face challenges in obtaining the layout of structures, such as buildings, when this information is not available, and existing methods struggle to effectively filter out reverberation data from sensor images, making it difficult to identify structural features.
Innovation Solution
A method and system that filter sensor data to distinguish between edge data and reverberation data, using a computing system with image generator logic to produce a filtered image from edge data only, by identifying and separating sensor data into bins and comparing intensities to determine mathematically-determined thresholds for edge and reverberation data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If reverberation data is included in sensor images, then more complete structural information is captured, but structural features become harder to identify due to noise interference
Solution Approach 1:
The patent segments sensor data into multiple bins representing different spatial locations, then applies intensity comparison logic to distinguish edge data from reverberation data in each bin. This segmentation allows selective processing of different data components to resolve the contradiction between information completeness and identification accuracy.
Solution Approach 2:
The patent extracts and removes reverberation data from the sensor image by comparing intensities across bins and identifying characteristic reverberation patterns. This extraction process eliminates the harmful reverberation component while preserving the useful edge data, thereby resolving the contradiction.
2Object-affected harmful factors
If traditional filtering methods are used to remove reverberation data, then noise reduction is achieved, but structural features may also be lost or distorted
Solution Approach 1:
The patent employs feedback mechanisms where the intensity comparison logic continuously evaluates bin intensities and adjusts the filtering decision based on the detected patterns. This feedback approach allows the system to distinguish between reverberation and actual structural features, removing harmful noise while preserving reliable structural information.
Solution Approach 2:
The patent changes the parameter evaluation criteria by comparing intensities across multiple bins and identifying specific intensity relationship patterns that characterize reverberation. This parameter-based discrimination allows selective removal of reverberation data while preserving structural feature data, resolving the contradiction between noise reduction and feature preservation.
3Productivity
If all sensor data is used to generate images, then processing speed is maintained, but image clarity deteriorates due to reverberation interference
Solution Approach 1:
The patent performs preliminary intensity comparison and reverberation identification on binned sensor data before final image generation. This preliminary action filters out reverberation data in advance, so that the subsequent image generation process uses only clean edge data, achieving both speed and clarity.
Solution Approach 2:
By segmenting sensor data into bins and processing each bin through intensity comparison logic, the patent enables parallel processing of multiple data segments. This segmentation approach maintains overall processing speed while improving image clarity through selective reverberation removal in each segment.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
According to one embodiment, a method comprises receiving sensor data generated by one or more sensors in response to sensing a structure. The sensor data is filtered to identify edge data and reverberation data each describing the same structural feature of the structure. Image data for a filtered image of the structure is generated from the edge data, but not from the reverberation data.