Image Noise Reduction Through PSF Modulation and ROIC Demodulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional filtering techniques for infrared signals are susceptible to interference and noise, leading to non-optimal performance in real-world scenarios due to environmental factors and limitations in existing filtering mechanisms.
Innovation Solution
Implementing a modulation and demodulation technique using a Computational Pixel Imager (CPI) device that modulates optical signals at a frequency higher than the excess noise knee frequency and demodulates them within each pixel, combined with digital signal processing to filter out unwanted noise, such as 1/fn noise, using a read-out integrated circuit (ROIC).
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional filtering techniques are used for infrared signals, then the filtering mechanism is simple, but the signal clarity and performance are degraded due to interference and noise
Solution Approach 1:
The patent applies dynamics by modulating the optical signal at a high frequency (e.g., 100 Hz to 10 kHz) to shift the signal spectrum away from low-frequency noise. This dynamic frequency modulation allows the signal to be separated from noise in the frequency domain, improving signal clarity without requiring complex spatial filtering mechanisms
Solution Approach 2:
The patent uses periodic modulation of the optical signal at a defined frequency, and corresponds periodic demodulation at the same frequency. This periodic action creates distinct frequency signatures that enable easy separation of signal from noise through frequency-selective filtering, achieving high signal clarity with relatively simple filtering circuits
Solution Approach 3:
The patent introduces an intermediary modulation frequency as a mediator between the optical signal and the noise. By embedding the signal at a frequency远高于the excess noise knee frequency, the modulation acts as an intermediary that separates the signal from low-frequency noise sources, allowing simple filters to effectively remove noise while preserving signal integrity
2Reliability
If modulation and demodulation technique is implemented, then the signal-to-noise ratio is improved, but the device complexity increases due to additional modulation components
Solution Approach 1:
The patent achieves multi-functionality by integrating modulation and demodulation capabilities into the existing optical detection system. The modulation can be implemented through various means (electro-optic modulators, acoustic-optic modulators, or direct modulation of light sources), and the demodulation can be performed in the optical domain or electrical domain, allowing the same basic architecture to serve multiple application scenarios with different complexity requirements
Solution Approach 2:
The patent replaces complex mechanical filtering systems with frequency-domain separation achieved through modulation. Instead of using complex mechanical or spatial filters to separate signal from noise, the system modulates the signal to a higher frequency where simple electronic or optical filters can effectively reject low-frequency noise, substituting mechanical complexity with frequency-domain processing
3Ease of manufacture
If traditional Non-Uniformity Correction is used, then the processing is simple, but the noise reduction performance is insufficient in real-world scenarios
Solution Approach 1:
The patent applies preliminary action by performing modulation of the optical signal before detection and conversion to electrical signals. This pre-modulation shifts the signal spectrum to high frequencies before any noise contamination occurs in subsequent stages, allowing noise reduction to be achieved through frequency-selective filtering rather than requiring complex post-processing correction algorithms
Data Source
AI summary
A system, method and device for filtering noise from image data is disclosed. The method includes: (a) obtaining an input optical signal input to an optical system, (b) modulating the input optical signal to obtain a modulated image signal, wherein the input optical signal is modulated based at least in part on manipulating a point spread function (PSF) of the optical system, (c) providing the modulated image signal to a read-out integrated circuit (ROIC), and (d) obtaining a filtered image signal from the ROIC, wherein the ROIC filtered the modulated image signal to remove at least part of excess noise.