Computer Vision Sensor Using Color-Opponent Pattern
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Computer vision (CV) applications on mobile devices face challenges due to high power and processing usage from color image sensors, while limiting CV to grayscale reduces effectiveness, and hyperspectral light offers untapped functionality.
Innovation Solution
A CV image sensor with a repeating pattern of two pairs of color-opponent color sensors and a luminance sensor, optimizing surface area, hyperspectral sensitivity, and low-light capabilities, which sidesteps processor-intensive color conversions and enhances functionality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional color image sensors are used, then color detection functionality is provided, but power consumption and processing usage increase
Solution Approach 1:
The sensor array is segmented into multiple sensor types (luminance sensors and color sensors) arranged in a repeating pattern, where each sensor type performs a specific function. This segmentation allows the system to process different aspects of light independently, reducing the need for intensive post-processing color conversions while maintaining color detection capability.
Solution Approach 2:
The patent extracts the color conversion processing step from the imaging pipeline by having color-opponent sensors directly output color-opponent values. This eliminates the need for processor-intensive RGB color conversions, thereby reducing power consumption while preserving color detection functionality.
2Use of energy by moving object
If grayscale images are used, then power consumption is reduced, but effectiveness of color-dependent CV applications is limited
Solution Approach 1:
The sensor array dynamically adapts to different lighting conditions and application requirements by incorporating both luminance sensors (for low-light performance) and color sensors (for color detection). The system can adjust its operational characteristics based on scene luminance and color content, providing color detection capability when needed while maintaining energy efficiency.
3Adaptability or versatility
If RGBG pattern is used, then color detection is achieved, but processor-intensive color conversions are required
Solution Approach 1:
The patent removes the color conversion processing step by having color-opponent sensors directly measure color-opponent values (such as red-green and blue-yellow opponent channels). This extraction of the color conversion function from the processing pipeline eliminates the need for computationally intensive RGB to other color space conversions, thereby reducing device complexity.
Solution Approach 2:
The color-opponent sensors perform the color measurement function directly at the sensor level, providing color-opponent values without requiring external processing. This self-service approach at the sensor level eliminates the need for complex post-processing color conversions, reducing overall system complexity.
4Measurement precision
If more color sensors are added, then color detection accuracy improves, but surface area and power requirements increase
Solution Approach 1:
Each sensor in the array is designed to be multi-functional, with sensors capable of detecting both luminance and color information, or color-opponent values. This universality allows the system to achieve accurate color detection with fewer dedicated color sensors, thereby reducing the overall surface area requirement while maintaining measurement precision.
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 configuration maintains low power requirements while providing high functionality, increased light sensitivity, and efficient color and luminance detection, addressing the limitations of traditional RGBG patterns and enabling effective color detection without intensive processing.
Implementation Method 1
each color sensor is configured to detect a unique respective color of light. The four color sensors comprise a first pair of color-opponent color sensors configured to detect light from a first pair of opponent colors, and a second pair of color-opponent color sensors configured to detect light from a second pair of opponent colors
Implementation Method 2
a third value indicative of light detected by the luminance sensor
Data Source
AI summary
Embodiments described herein can address these and other issues by providing a CV image sensor that has a repeating pattern of two pairs of color-opponent color sensors and a luminance sensor to provide a high amount of functionality while maintaining relatively low power requirements. By using color opponency, embodiments can sidestep the need for processor-intensive color conversions required of other color sensors. Embodiments may optionally provide for surface area optimization, hyperspectral sensitivity, low-light optimization, neuromorphic light sensing, and more.