16-bit Floating-Point Image Sensor Dynamic Range Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current imaging systems, particularly CCD-based ones, face limitations in dynamic range, power consumption, and integration capabilities, making them unsuitable for many applications, while CMOS image sensors struggle with wide dynamic range imaging due to issues like fixed pattern noise and complex reconstruction processes.
Innovation Solution
A 16-bit floating-point format processor that manages the exponent in the analog domain and the mantissa in the digital domain, mimicking the human eye's dynamic range and representational precision, allowing for efficient capture and processing of wide dynamic range images with reduced power dissipation and cost.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If CCD-based image sensors are used, then image quality and dynamic range are improved, but power consumption increases and integration capability deteriorates
Solution Approach 1:
The patent replaces the mechanical/electrical charge transfer mechanism of CCDs with a CMOS-based electronic system that uses voltage domain sampling and digital signal processing to achieve wide dynamic range imaging, thereby reducing power consumption while maintaining integration capability
Solution Approach 2:
The patent changes the operating parameters by using voltage domain sampling with multiple exposure levels and logarithmic compression to achieve wide dynamic range in CMOS sensors, avoiding the high power consumption of CCDs while maintaining image quality
2Measurement precision
If CCD-based image sensors are used, then image quality is improved, but device complexity and fabrication cost increase
Solution Approach 1:
The patent merges the light sensing function with digital signal processing functions into a single CMOS integrated circuit, eliminating the need for separate support chips required by CCDs and reducing overall device complexity
Solution Approach 2:
The CMOS sensor design integrates multiple functions including light detection, voltage conversion, analog-to-digital conversion, and image processing within the same chip, making the system more versatile and easier to manufacture
3Measurement precision
If logarithmic response CMOS image sensors are used, then dynamic range is improved, but fixed pattern noise increases
Solution Approach 1:
The patent introduces an intermediary processing stage using voltage domain sampling with multiple exposure levels and digital reconstruction algorithms that mediate between the logarithmic response and the final image output, reducing fixed pattern noise while preserving dynamic range
Solution Approach 2:
The patent uses multiple frame capture techniques where multiple copies of the image at different exposure levels are taken and then digitally reconstructed, allowing the fixed pattern noise to be differentiated from actual image content and removed
4Measurement precision
If multiple frame capture techniques are used, then dynamic range is improved, but processing complexity increases
Solution Approach 1:
The patent performs preliminary actions by capturing multiple frames at different exposure levels before the reconstruction process, allowing the complex reconstruction algorithms to work with pre-processed data that has already been optimized for dynamic range combination
Data Source
AI summary
Embodiments of the invention provide a 16 bit floating point signal processor will typically give an order of magnitude more performance than a 32 bit floating point signal processor and about twice as much performance as a 16 bit fixed point processor. Capturing wide dynamic range images in 16 bit floating point format entails representing an iris of a imaging device as an exponent of a floating point number and representing the precision of said imaging device as a mantissa of said floating point number.