Endoscope Pixel Array Noise Correction via Dummy Signal Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Endoscopes using solid state image sensors often experience vertical stripe noise due to variations in element characteristics, which existing techniques fail to adequately address through effective pixel area correction methods.
Innovation Solution
The endoscope system incorporates a two-dimensional matrix of pixels with adjacent pixels sharing vertical lines, dummy pixels for signal correction, a threshold recording unit, a determination unit, a correction data generator, and a correction unit to calculate and apply correction data for imaging signals based on dummy signal statistics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If correction data is generated using all dummy signal output values, then correction processing is simple, but measurement precision deteriorates due to inclusion of abnormal values
Solution Approach 1:
The determination unit performs preliminary filtering of dummy signal output values before correction data generation. By determining in advance whether each output value is within the threshold range, the system prepares clean data for statistical processing, ensuring that only valid measurements contribute to the correction data.
Solution Approach 2:
The system uses the threshold range as a feedback criterion to evaluate dummy signal output values. The determination unit compares each output value against the threshold and selectively includes or excludes it based on this feedback, creating a quality-controlled dataset for correction data generation.
2Measurement precision
If threshold range is set narrowly to ensure precision, then measurement precision improves, but loss of information increases due to exclusion of potentially valid data
Solution Approach 1:
The threshold range parameter is dynamically adjusted based on the statistical properties of dummy signal output values. By calculating the standard deviation and setting the threshold as mean ± k×standard deviation, the system adapts the acceptance criterion to the actual data distribution, balancing precision and information retention.
Solution Approach 2:
The system applies a threshold-based filtering mechanism that selectively processes dummy signal values. By using a statistically derived threshold rather than an arbitrary fixed value, the system performs partial filtering that removes only clearly abnormal values while preserving the majority of valid data points.
3Reliability
If multiple dummy pixels are used for correction, then reliability improves through increased data samples, but device complexity increases due to additional components
Solution Approach 1:
The dummy pixels are integrated into the regular pixel array structure and share the same readout circuitry and data processing pathways. This multi-functional design allows dummy pixels to contribute to correction data generation without requiring separate dedicated hardware systems, thereby improving reliability while minimizing additional complexity.
Solution Approach 2:
The correction data generation unit combines data from multiple dummy pixels through statistical processing. By merging the output values from multiple dummy pixels and calculating their mean and standard deviation, the system leverages the combined information to produce more reliable correction data while using shared processing resources.
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 effectively corrects imaging signals by generating and applying correction data, reducing noise and improving image quality by using dummy signals to determine and adjust for variations in pixel output values.
Implementation Method 1
Solid state image sensors each having a plurality of pixels are configured to receive light and perform photoelectric conversion on the light and output an electrical signal
Data Source
AI summary
An endoscope includes: pixels arranged in a matrix form and configured to generate an imaging signal according to an amount of received light, and output the imaging signal through one of vertical lines in the pixels, each two pixels adjacent to one another in a horizontal direction sharing a single vertical line; dummy pixels, each provided for each vertical line in the pixels; a determination unit for determining whether output values of dummy signals having been output multiple times from a dummy pixel of the dummy pixels, are within a range of a threshold; a correction data generator configured to: calculate, for each vertical line, a statistic of the output values of the dummy signals determined to be within the range of the threshold; and generate, for each vertical line, correction data based on the calculation; and a correction unit for correcting the imaging signal based on the correction data.