Linear Sensor Asymmetric Pixels OCT Alignment

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In optical coherence tomography (OCT) imaging, precise alignment of the image line with respect to the linear sensor is crucial for capturing accurate spectral information, but existing methods rely on manual adjustments and are prone to misalignment, leading to loss of information and reduced image quality due to the tall pixels used for alignment, which can be sensitive to vibrations and temperature changes.

Innovation Solution

The implementation of a linear sensor with pairs of detection pixels that provide asymmetrical responses to misalignment, allowing for precise determination of image line positioning and correction, using optical masks to differentiate between image pickup and alignment detection pixels, and potentially placing detection pairs at both ends and the middle of the sensor line.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual alignment adjustment methods are used, then device complexity is reduced, but alignment precision deteriorates leading to information loss

Engineering Contradiction:
Improvealignment precisionVSAvoiddevice complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The linear sensor performs self-alignment detection by using its own pixel outputs to determine misalignment. The detection pixels automatically provide alignment information without requiring external alignment tools or complex adjustment mechanisms, enabling the system to self-correct alignment issues through software processing of pixel signal differences.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical alignment adjustment with an optical-electrical detection system. Instead of physically adjusting components to achieve alignment, the system uses detection pixels to optically sense the image line position and electronically process the signal differences to determine alignment status, substituting mechanical adjustment with optical detection and electronic control.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

2Measurement precision

If tall pixels are used for alignment detection, then alignment sensitivity is improved, but reliability deteriorates due to sensitivity to vibrations and temperature changes

Engineering Contradiction:
Improvealignment sensitivityVSAvoidreliability
Core Design Contradiction:
Measurement precisionVSReliability

Solution Approach 1:

The patent introduces asymmetry in the pixel response characteristics to enable alignment detection. Detection pixels are designed with different effective photosensitive surface widths that vary monotonically in opposite directions, creating asymmetric responses to misalignment. This asymmetry allows the system to detect alignment status through signal differences while maintaining stable, vibration-resistant pixel structures.

Inventive Principle:
Principle #4Asymmetry

Solution Approach 2:

The patent applies local quality differentiation by giving detection pixels specialized photosensitive surface characteristics distinct from regular imaging pixels. The detection pixels have asymmetric effective photosensitive widths optimized for alignment detection, while other pixels maintain uniform characteristics for imaging, allowing each region to have the quality needed for its specific function.

Inventive Principle:
Principle #3Local quality

3Manufacturing precision

If optical masks are used to differentiate detection pixels, then manufacturing precision is improved, but device complexity increases

Engineering Contradiction:
Improvepixel differentiation precisionVSAvoiddevice complexity
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The patent performs preliminary action by pre-configuring the effective photosensitive surface widths of detection pixels during manufacturing. The asymmetric photosensitive surfaces are built into the pixel structure or defined by optical masks during fabrication, allowing the pixels to inherently provide alignment detection functionality without requiring additional runtime components or complex control mechanisms.

Inventive Principle:
Principle #10Preliminary action

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 approach enables automatic alignment correction, reducing information loss and improving image quality by providing precise alignment information directly from sensor outputs, enhancing the reliability and precision of OCT imaging across various conditions.

Implementation Method 1

A spectrometer 20 comprises a dispersive element 21, for example a diffraction grating associated with optical elements (lenses), which disperses the spectrum of light received from the output arm of the interferometer as a function of the wavelength

Methodology Applied
Scientific EffectDiffraction: Diffraction

Implementation Method 2

The linear sensor 31 comprises a line of photosensitive pixels, for example 1024 pixels or 2048 pixels, and the sequencing circuitry for image capture and pixel reading. Each pixel makes it possible to measure the light intensity for a determined wavelength of the spectrum.

Methodology Applied
Scientific EffectPhotoelectric effect: Photoelectric Effect

Data Source

PatentEP3479407B1Linear sensor for a spectroscopic optical coherence tomography imaging apparatus
Publication Date: 2021.09.01 TELEDYNE E2V SEMICON SAS
  • EP3479407B1 patent drawingFigure 1
  • EP3479407B1 patent drawingFigure 2a~2c
  • EP3479407B1 patent drawingFigure 3

AI summary

The invention provides a linear sensor that includes a row of N pixels 100 that are regularly spaced apart and formed in a semiconductor substrate, and a circuit for reading the N pixels, which delivers an output signal for each of the N pixels of the row, the sensor being characterised in that the N pixels comprise image-taking pixels that have a rectangular-shaped useful photosensitive area R of greater height than width, where width is the dimension in the direction of the row and height the dimension in the perpendicular direction, and at least two pairs of alignment-detecting pixels PPL, PPR, which pairs are separated by a distance, the detecting pixels of each pair being adjacent pixels P1,P2 in the row of pixels 100 and the useful photosensitive area DT1, DT2 of each of the detecting pixels of a given pair having a width that varies monotonically in the height direction, one from top to bottom and the other from bottom to top.