Image Sensor Depth Pixel Memory Isolation Walls
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image sensors that capture both 2D and depth information face challenges in alignment and integration due to the mismatched dimensions of depth and 2D image pixels, leading to misalignment and increased volume and cost when using separate sensors or integrating them within a single array.
Innovation Solution
An image sensor design that incorporates depth pixels with multiple memory zones acting as isolation walls between 2D image pixels, allowing for efficient integration and alignment of both pixel types within a single array, with each depth pixel having a detection zone and multiple memories forming partial isolation walls between adjacent 2D image pixels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If separate image sensors are used to capture 2D image and depth image, then both images can be captured, but misalignment between pixels occurs and volume and cost increase
Solution Approach 1:
The patent combines both 2D image pixels and depth pixels within a single sensor array, eliminating the need for separate sensors. This merging approach ensures that both image types are captured from the same viewpoint, preventing misalignment while reducing overall sensor volume and cost.
Solution Approach 2:
The sensor array is designed to perform multiple functions: capturing both 2D images and depth images simultaneously using the same physical sensor structure. This multi-functionality resolves the contradiction by allowing one sensor to replace what would traditionally require two separate sensors.
2Reliability
If separate image sensors are used to capture 2D image and depth image, then both images can be captured, but device cost increases
Solution Approach 1:
The patent combines both 2D image pixels and depth pixels within a single sensor array, eliminating the need for separate sensors. This merging approach ensures that both image types are captured from the same viewpoint, preventing misalignment while reducing overall sensor volume and cost.
3Reliability
If both 2D image pixels and depth pixels are integrated within a same sensor array, then alignment is improved, but integration becomes complex due to dimension mismatch
Solution Approach 1:
The patent resolves the dimensionality conflict by extending memory structures vertically into the third dimension. The first, second, and third memories extend to form isolation walls between adjacent 2D image pixels, allowing depth pixels (which have larger dimensions) to be integrated with 2D pixels without increasing planar complexity.
Solution Approach 2:
The patent segments the sensor array into distinct pixel types (2D image pixels and depth pixels) that can be independently designed and optimized. Each pixel type has its own structure characteristics, allowing them to coexist in the same array without requiring uniform dimensions across all pixels.
4Measurement precision
If depth pixels have larger dimensions than 2D image pixels, then depth detection capability is improved, but integration with 2D pixels becomes complex
Solution Approach 1:
The patent resolves the dimensionality conflict by extending memory structures vertically into the third dimension. The first, second, and third memories extend to form isolation walls between adjacent 2D image pixels, allowing depth pixels (which have larger dimensions) to be integrated with 2D pixels without increasing planar 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 design enables the capture of both 2D and depth images with improved alignment and reduced size and cost, while maintaining effective charge storage and isolation between pixels, facilitating the calculation of depth information using phase shift detection of light signals.
Implementation Method 1
a doped region sandwiched between first and second parallel straight walls, the first and second walls of each memory having a conductive core adapted to receive a biasing voltage
Data Source
AI summary
The invention concerns an image sensor comprising: a depth pixel (PZ) having: a detection zone (PD); a first memory (mem1) electrically coupled to the detection zone by a first gate (310); a second memory (mem2) electrically coupled to the detection zone by a second gate (314); and a third memory (mem3) electrically coupled to the detection zone by a third gate (316), wherein the first, second and third memories are each formed by a doped region sandwiched between first and second parallel straight walls (404, 406), the first and second walls of each memory having a conductive core adapted to receive a biasing voltage; and a plurality of 2D image pixels (P1 to P8) positioned adjacent to the depth pixel, wherein the first, second and third memories extend to form at least partial isolation walls between corresponding adjacent pairs of the 2D image pixels.