Compressive Sensing Image Feature Recovery Without Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for identifying geometric features in signals require reconstructing the entire signal, which is energy-intensive and computationally expensive, and often result in ambiguous target recovery in optical superposition methods.
Innovation Solution
The proposed compressive sensing system efficiently acquires and recovers geometric features by folding an image into compressed representations, allowing for feature recovery without reconstructing the signal, using methods such as hashing functions and error correcting codes, reducing the number of measurements required and enabling sub-linear algorithmic complexity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If uniform sampling based methods are used to acquire the entire signal, then the signal can be fully captured, but significant energy is expended even though only small parts of the signal contain the features of interest
Solution Approach 1:
The patent extracts only the relevant feature information from the signal rather than acquiring the entire signal. By using compressive sensing techniques, the system directly measures and recovers features of interest (such as corners, edges, or specific geometric properties) without capturing unnecessary signal portions, thereby reducing energy consumption while maintaining feature identification accuracy.
Solution Approach 2:
The patent applies partial action by acquiring only the minimum necessary measurements required to identify features of interest. Instead of uniform sampling that captures the entire signal, the system performs partial sampling that targets specifically the information needed for feature detection, reducing energy expenditure on redundant data acquisition.
2Measurement precision
If standard compressive sensing based approaches are used to recover the entire signal, then the signal can be reconstructed, but a computationally expensive algorithm is required and the recovered signal is then used as an input to feature identification
Solution Approach 1:
The patent extracts feature information directly from compressive measurements without performing full signal reconstruction. By formulating feature identification as a direct optimization problem on the compressive measurements, the system eliminates the computationally expensive reconstruction step while still achieving accurate feature detection.
Solution Approach 2:
The patent performs partial reconstruction by directly recovering only the feature parameters of interest from compressive measurements rather than reconstructing the entire signal. This partial action approach significantly reduces computational complexity by focusing only on the essential information needed for feature identification.
3Adaptability or versatility
If optical superposition or multiplexing methods are used to superimpose different parts of a scene, then wide field of view cameras can be enabled for target detection and tracking, but the recovery of the target position in the original scene is ambiguous and requires additional knowledge about the target object
Solution Approach 1:
The patent introduces an intermediary computational step that resolves the ambiguity in target position recovery. By using compressive sensing techniques with appropriately designed measurement matrices, the system acts as an intermediary that maps the superimposed optical information back to unique target positions, eliminating the ambiguity without requiring additional knowledge about the target object.
Data Source
AI summary
Methods and apparatuses for compressive sensing that enable efficient recovery of features in an input signal based on acquiring a few measurements corresponding to the input signal. One method of compressive sensing includes folding an image to generate first and second folds, and recovering a feature of the image based on the first and second folds without reconstructing the image. One example of a compressive sensing apparatus includes a lens, a focal plane array coupled to the lens and configured to generate first and second folds based on the image, and a decoder configured to receive the first and second folds and to recover a feature of the image without reconstructing the image. The feature may be a local geometric feature or a corner. Compressive sensing methods and apparatuses for determining translation and rotation between two images are also disclosed.