Smart CMOS Bright-Pixel Readout for High-Speed 3D Reconstruction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional image sensors output whole images, including useless dark pixel information, leading to high bandwidth requirements and low readout speed, particularly in 3D laser scanning where only bright pixel locations and intensities are needed.
Innovation Solution
A novel smart CMOS image sensor selects bright pixels based on thresholding criteria, outputs their intensities and locations only, uses parallel I/Os for data export, employs re-routing and memory buffers to distribute data evenly, and applies interleaved timing to reduce bandwidth and speed up ADC conversion.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional image sensors output whole images, then complete image information is captured, but bandwidth requirements increase and readout speed decreases
Solution Approach 1:
The patent extracts only the necessary bright pixel information (intensity and location) from the complete image data, eliminating unnecessary dark pixel data. This extraction is achieved through on-chip pixel selection circuitry that identifies and outputs only bright pixels, directly reducing bandwidth requirements while maintaining the completeness of useful information for 3D reconstruction
Solution Approach 2:
The patent segments the image data by separating bright pixels from dark pixels and processing them differently. The pixel array is divided into selectable pixels based on intensity thresholds, allowing independent processing and output of only the relevant bright pixel data, thereby reducing the overall data volume without losing critical information
2Loss of information
If traditional image sensors output whole images, then complete image information is captured, but bandwidth requirements increase
Solution Approach 1:
The patent extracts only the necessary bright pixel information (intensity and location) from the complete image data, eliminating unnecessary dark pixel data. This extraction is achieved through on-chip pixel selection circuitry that identifies and outputs only bright pixels, directly reducing bandwidth requirements while maintaining the completeness of useful information for 3D reconstruction
Solution Approach 2:
The patent applies partial action by processing and outputting only a subset of pixels (bright pixels) rather than the entire image array. The pixel selection circuitry selectively processes only those pixels exceeding intensity thresholds, reducing data volume while maintaining sufficient information for 3D shape reconstruction
3Quantity of substance
If pixel selection is implemented, then bandwidth is reduced, but device complexity increases
Solution Approach 1:
The patent merges the pixel selection function with the existing CMOS image sensor architecture by integrating selection circuitry at the pixel level. The pixel array itself serves as both the light sensor and the selection mechanism, where each pixel's intensity is directly compared to a threshold, combining multiple functions into a unified structure that reduces overall system complexity
Solution Approach 2:
The pixel selection circuitry operates autonomously at the pixel level, where each pixel automatically compares its intensity to a threshold and selects itself for output if it meets the criteria. This self-service mechanism eliminates the need for complex external processing systems, reducing device complexity while maintaining effective pixel selection
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 reduces bandwidth requirements, increases readout speed, and enables high-frame-rate 3D reconstruction by focusing on selected pixel data, facilitating efficient 3D scanning and object tracking.
Implementation Method 1
a pixel array for converting light intensity to electrical signals
Data Source
AI summary
This disclosure presents a novel smart complementary metal oxide semiconductor (CMOS) sensor that can detect the “bright” pixels and export the light intensity and location of the selected pixels only. The detecting function is achieved by applying a thresholding criteria method. A novel CMOS architecture is proposed. The FPA-implemented COMS is shared by two sets of column processing circuitry for selecting, processing and exporting the data from top half and bottom half of FPA respectively. The CMOS architecture comprises a re-routing scheme, multiple I/Os deployment, parallel-shifting FIFO memory buffers, and interleaved timing scheme.