Compound Eye Image Sensing with Sparse Representation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image sensing apparatuses using compound eyes struggle to recover high-resolution images when the distance between the object and the apparatus is unknown, as they require specific measurement matrices for accurate image processing.
Innovation Solution
An image sensing apparatus equipped with compound eyes, memory storing multiple measurement matrices for different distances, and digital signal processing techniques for sparse representation, allowing for image recovery and distance measurement using light amount information and residual norms or energy comparison.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the number of lenses is increased to improve image resolution, then image resolution is improved, but device complexity and space requirements increase
Solution Approach 1:
The image sensing apparatus divides the compound eye into multiple simple eyes (lenses), where each lens captures light from a specific direction. The sensed image from each lens is processed independently through compression sensing techniques, allowing high-resolution image recovery without requiring an excessive number of lenses. This segmentation approach enables resolution improvement while controlling device complexity.
Solution Approach 2:
The patent arranges multiple lenses on a three-dimensional curved surface rather than a flat plane, utilizing spatial dimensionality to pack more lenses efficiently. This 3D arrangement maximizes the field of view and light capture capability within a compact form factor, improving image resolution without proportionally increasing device complexity.
2Measurement precision
If multiple measurement matrices are stored for different distances, then image recovery accuracy is improved, but memory requirements and device complexity increase
Solution Approach 1:
The system pre-calculates and stores multiple measurement matrices corresponding to different object distances before actual image sensing. These measurement matrices are prepared in advance and stored in memory, enabling rapid image recovery without real-time calculation. This preliminary action ensures accurate image recovery for objects at various distances while avoiding complex real-time computations.
Solution Approach 2:
The patent changes the parameter of measurement matrix selection based on the distance to the object. By storing measurement matrices with different distance parameters and selecting the appropriate matrix based on measured distance, the system achieves accurate image recovery for various object distances without storing excessive data for all possible distances.
3Measurement precision
If compression sensing technique is used to recover images, then image recovery accuracy is improved, but calculation complexity and processing time increase
Solution Approach 1:
The image recovery process uses feedback mechanisms where the sensed image from each lens is iteratively processed through compression sensing algorithms. The system refines the recovered image by comparing expected patterns with actual sensed data, adjusting the reconstruction until convergence. This feedback-based approach improves image recovery accuracy while managing calculation complexity through iterative optimization.
Data Source
AI summary
There is provided an image sensing apparatus according to the present invention including: an image recovery portion which recovers an image from a sensed image by using a digital signal processing technique which uses a sparse representation for each measurement matrix; and a distance measurement portion which measures a distance of an object by using a norm of a value using at least light amount information of the estimated pixel which is estimated for each measurement matrix, in which the image recovery portion recovers the image by using the measurement matrix corresponding to the distance measured by the distance measurement portion. According to the present invention, when an image is sensed using compound eyes, image sensing can be performed without a separate distance measuring device or the like.