Multi-Spectral Camera-on-Chip Layout for Alignment-Free Image Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional multi-camera systems are costly, large, consume high power, generate significant data, and require complex camera alignment and calibration, especially when dealing with different fields of view and functionalities, which can be disrupted by vibrations and physical distortions.
Innovation Solution
A multi-camera system integrated on a single sensor chip with shared main lenses, metalenses, and microlenses, where each sensor array covers the same field of view and includes nanostructures to split and direct different light spectrums, allowing for on-chip image processing and fusion without the need for external calibration.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple cameras with different fields of view and functionalities are used, then imaging capabilities are improved, but system cost, size, and power consumption increase
Solution Approach 1:
The patent combines multiple sensor arrays (RGB, NIR, SWIR, UV, TOF) onto a single sensor chip, sharing common optical elements (main lens, metalens, microlens arrays) and processing resources. This integration reduces the total number of separate camera modules, thereby lowering power consumption while maintaining diverse imaging capabilities across different spectral ranges and functionalities.
Solution Approach 2:
The patent implements a universal optical path where a single main lens and metalens serve multiple sensor arrays simultaneously. The shared optical elements can focus different wavelengths of light to respective sensor arrays, enabling one optical system to perform multiple imaging functions (visible, infrared, ultraviolet, time-of-flight) without requiring separate dedicated optics for each sensor type.
2Adaptability or versatility
If multiple cameras with different fields of view are used, then functional versatility is improved, but device complexity and alignment requirements increase
Solution Approach 1:
By merging multiple sensor arrays onto a single chip with shared optical elements, the patent eliminates the need for complex external alignment between separate camera modules. The unified structure ensures precise alignment of optical paths from the main lens and metalens to each sensor array, reducing manufacturing complexity while maintaining versatile functional capabilities.
3Ease of manufacture
If cameras are aligned externally, then initial setup is possible, but alignment is easily disturbed by vibration and physical distortion
Solution Approach 1:
The patent integrates multiple sensor arrays and their associated microlens arrays onto a single sensor chip, creating a rigid, monolithic structure. This integration ensures that relative positions and alignments between optical elements and sensors are fixed at the semiconductor manufacturing level, making the system inherently resistant to vibration and physical distortion that would affect externally aligned separate camera modules.
4Adaptability or versatility
If camera calibration is performed externally, then image fusion is enabled, but processing power and time are consumed
Solution Approach 1:
The patent performs calibration and alignment during the semiconductor manufacturing process itself, before the device is deployed. The precise positioning of sensor arrays and optical elements is established through fabrication techniques, eliminating the need for time-consuming external calibration procedures. This preliminary action ensures that image fusion can be performed efficiently without requiring extensive post-manufacturing adjustment or real-time calibration under vibrating conditions.
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 solution reduces power consumption, eliminates alignment issues, and enables real-time image processing and fusion on a single chip, improving system efficiency and reducing latency by integrating multiple cameras with the same field of view and functionalities.
Implementation Method 1
The metalens includes at least one layer of nanostructures formed on a substrate in which the at least one layer of nanostructures diffracts and/or focuses at least one spectrum of the light incident on the metalens to the corresponding sensor array
Implementation Method 2
Nanostructures of the at least one layer of nanostructures include a material having a refractive index greater than 1.9
Implementation Method 3
The optical path includes a main lens, a metalens and two or more microlens arrays. The main lens and the metalens are shared by each sensor array
Implementation Method 4
at least one sensor array may include a single-photon avalanche diode
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
A camera system includes two or more sensor arrays and an optical path. The sensor arrays are on the same sensor chip. Each sensor array includes the same field of view (FOV) as each other sensor array. The optical path includes a main lens and a metalens that are shared by each sensor array, and a microlens associated with each sensor array. The metalens splits incident light into different spectrums of light and directs each respective spectrum to a corresponding sensor array. The different spectrums of light include at least two of visible light, near infrared light, shortwave infrared and longwave infrared, and at least one sensor array includes single-photon avalanche diodes. The image processor that provides image processing, object recognition and object tracking and/or image fusion functionality may be on the same sensor chip as the sensor arrays.