Pixel Histogram Bin Mapping for Memory-Efficient SPAD Ranging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional imaging systems using single photon avalanche diodes (SPADs) face challenges in achieving small pixel sizes while maintaining high temporal resolution and range resolution due to the significant memory requirements for storing time-of-arrival histogram data, which occupy a large percentage of pixel real estate.
Innovation Solution
The proposed solution involves dividing the frame collection time into multiple collection subframes, using different bin mappings and potentially different bin widths for each subframe, allowing for a reduced number of bins and memory requirements without compromising detection range or resolution, and employing techniques like bin offsetting to ensure unique bin combinations for range determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If in-pixel memory is used to store time-of-arrival histogram data, then power consumption is reduced and signaling area is minimized, but pixel size increases significantly
Solution Approach 1:
The frame collection time is divided into multiple collection subframes, and the histogram memory is segmented into multiple sets of bins. Different bin mappings are applied to different subframes, allowing the same physical memory to serve multiple temporal ranges. This segmentation enables reduced memory size per subframe while maintaining full frame coverage capability.
Solution Approach 2:
The patent introduces a temporal dimension to the memory usage by implementing time-wintering where different bin mappings are applied across different time subframes. This allows the memory to be reused across time, effectively multiplying its capacity without increasing physical size. The same memory locations store different histogram data for different subframes.
2Measurement precision
If the number of bins is increased to maintain detection range and resolution, then temporal and range resolution are improved, but memory requirements and pixel size increase
Solution Approach 1:
The bin mappings are made dynamic and configurable, allowing the system to adapt the number and width of bins based on the specific requirements of each collection subframe. This dynamic reconfiguration enables the memory to efficiently accommodate varying resolution requirements without being over-provisioned for all cases simultaneously.
Solution Approach 2:
The patent changes the parameters of the bin mappings between different collection subframes, including bin width, bin offset, and number of bins. By varying these parameters, the system optimizes memory usage for each subframe's specific detection requirements while maintaining overall detection range and resolution capabilities.
3Area of moving object
If multiple bin mappings are used for different collection subframes, then memory efficiency is improved and pixel size is reduced, but system complexity increases
Solution Approach 1:
The histogram memory and bin mapping circuitry are designed to be universal and multi-functional, serving multiple collection subframes with different bin mappings. This multi-functionality is achieved through configurable bin mapping logic that can be programmed with different mapping parameters, eliminating the need for separate dedicated memory for each subframe.
Solution Approach 2:
The patent uses copying of bin mapping configurations across different collection subframes, where the same physical memory structure is logically copied or replicated for different time periods with different mapping parameters. This allows complex multi-subframe operation without proportionally increasing hardware complexity.
4Measurement precision
If bin offsetting is implemented to ensure unique bin combinations for range determination, then range resolution is maintained, but memory access complexity increases
Solution Approach 1:
Bin offsets are pre-calculated and pre-configured for each collection subframe to ensure unique bin combinations. This preliminary action of pre-computing the offset values simplifies the runtime operation, as the memory access pattern is predetermined and can be efficiently implemented without complex real-time calculations.
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 pixel size and memory needs while maintaining high temporal and range resolution, enabling efficient data throughput by allowing a single readout of histogram data to resolve object ranges accurately.
Implementation Method 1
single photon avalanche diodes (SPADs) to support time-of-flight applications... the absorption of a single incident photon is designed to quickly produce an avalanche signal... Because the avalanche charge generated by the SPAD's breakdown grows so quickly
Data Source
AI summary
Techniques for resolving a range to an object using histograms are disclosed. A frame collection time for a depth-image frame is divided into a plurality of different collection subframes, where each collection subframe encompasses a plurality of light pulse cycles. Counts of accumulated photon detections by a pixel during the different collection subframes are allocated to histogram bins using different bin maps for the collection subframes. Each bin map defines a different mapping of time to bins for the light pulse cycles within its applicable collection subframe, and each mapping defines a bin width for its bins so that its bin map covers a maximum detection range for the depth-image frame. A range to an object in the pixel's field of view (within the maximum detection range) can be resolved according to a combination of peak bin positions in the histogram data with respect to the different collection subframes.