Light Receiving Pixel Filters for Optical Convolution Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Image recognition processing by convolution operations faces challenges in achieving high real-time performance due to the large amount of data to be processed and the complexity of the processing itself.
Innovation Solution
A light receiving device comprising first filters that transmit edge components in predetermined directions, second filters that transmit specific wavelength bands, and photoelectric conversion elements that convert light through these filters, enabling optical convolution operations for faster image recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image recognition processing is performed by convolution operation on image data, then recognition accuracy is improved, but processing speed deteriorates due to large data amount and complex processing
Solution Approach 1:
The patent replaces electronic/digital convolution operations with optical convolution operations. By using optical elements (lenses, mirrors, spatial light modulators) to perform the convolution mathematically in the optical domain, the system achieves parallel processing of all pixel data simultaneously, thereby maintaining high recognition accuracy while dramatically improving processing speed.
Solution Approach 2:
The patent transforms the convolution operation from the spatial domain to the frequency domain using optical Fourier transforms. This dimensional transformation allows the convolution operation to be performed as a simple multiplication in the frequency domain, reducing computational complexity from O(N²) to O(N) while preserving recognition accuracy.
2Productivity
If optical convolution operations are performed using filters for edge components and wavelength bands, then processing speed is improved, but device complexity increases
Solution Approach 1:
The patent designs optical elements that can perform multiple functions: the same optical convolution system can be reconfigured to perform different convolution operations by changing the spatial light modulator patterns or filter configurations. This multi-functionality reduces the need for multiple dedicated hardware components, thereby improving processing speed without proportionally increasing device complexity.
Solution Approach 2:
The patent employs programmable spatial light modulators and tunable filters that can dynamically change their characteristics during operation. This dynamic reconfigurability allows the system to adapt to different recognition tasks without requiring physical reconfiguration of the entire optical system, balancing processing speed improvement with manageable device complexity.
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
This configuration allows for higher-speed image recognition processing by performing convolution operations optically, potentially omitting the need for the first convolution layer in neural networks and enhancing processing speed and accuracy.
Implementation Method 1
a plurality of photoelectric conversion elements that each photoelectrically convert light transmitted through one of the plurality of first filters and one of the plurality of second filters
Data Source
AI summary
Higher-speed image recognition processing can be implemented. A light receiving device according to an embodiment includes: a plurality of first filters (130) each transmitting an edge component in a predetermined direction in an incident image; a plurality of second filters (150) each transmitting light of a predetermined wavelength band in incident light; and a plurality of photoelectric conversion elements (PD) each photoelectrically converting light transmitted through one of the plurality of convolution filters and one of the plurality of color filters.