Incoherent Noise Removal via Integral Transform
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for removing incoherent noise from marine geophysical data are expensive, time-consuming, and imprecise, often requiring iterative and parameter-dependent de-noising procedures that can either leave noise in the data or affect geophysical information.
Innovation Solution
The method involves subjecting the data set to an integral transform to create an identifiable pattern from incoherent noise events, modeling these patterns, and adaptively subtracting them using a least-squares error process, allowing for efficient removal of incoherent noise while preserving geophysical information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If conventional de-noising procedures are applied to remove incoherent noise, then noise reduction is achieved, but processing time and cost increase significantly
Solution Approach 1:
The patent transforms the data set from the time-domain to the frequency-domain using Fourier transform, changing the parameter domain to enable more efficient noise identification and removal. This parameter transformation allows noise events to be identified and removed in a single pass rather than requiring multiple iterative de-noising operations, significantly reducing processing time while maintaining effective noise reduction
Solution Approach 2:
The patent extracts individual noise events from the data set by identifying them as discrete entities in the frequency-domain. Each noise event is separated and removed independently through adaptive subtraction, allowing for efficient single-pass processing without the need for repeated iterative de-noising operations that characterize conventional methods
2Object-affected harmful factors
If conventional de-noising procedures are applied to remove incoherent noise, then noise reduction is achieved, but processing complexity increases due to iterative requirements
Solution Approach 1:
By transforming to the frequency-domain and changing the parameter representation of the data, the patent simplifies the identification and removal process. The transformation enables noise events to be clearly identified and removed in a single systematic pass, eliminating the need for complex iterative parameter tuning and multiple de-noising operations required by conventional methods
Solution Approach 2:
The patent employs an adaptive subtraction process that automatically identifies and removes noise events without requiring user intervention for parameter tuning or iterative quality control. The system self-adjusts to remove each noise event based on its identified characteristics, reducing processing complexity by eliminating the need for user-guided iterative procedures
3Object-affected harmful factors
If conventional de-noising procedures are applied to remove incoherent noise, then noise reduction is achieved, but geophysical information may be affected or lost
Solution Approach 1:
The patent extracts and removes only the identified noise events from the data set, leaving the remaining geophysical information intact. By treating noise as discrete extractable entities rather than applying broad de-noising filters, the method removes noise without affecting the underlying geophysical signals, preserving information that conventional aggressive de-noising might remove
Solution Approach 2:
The adaptive subtraction process uses feedback from the identified noise event characteristics to precisely target and remove only the noise components. This feedback mechanism ensures that removal operations are tailored to each specific noise event, preventing the loss of geophysical information that occurs with conservative parameterization in conventional methods
Data Source
AI summary
Noise is removed from a data set by performing an integral transform operation that converts instances of noise into identifiable artifacts in the transformed data set. A model of the artifacts is constructed by creating a full-domain or partial-domain noise model and performing the same integral transform operation on the noise model. The resulting transformation of the noise model is adaptively subtracted from the transformed data set to remove the noise. The adaptive subtraction may employ a least-square error filter.