Dual Execution Environment Image Processing for Face Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing technologies in intelligent terminals face challenges in efficiently acquiring and processing images for face recognition, particularly in ensuring data safety and processing speed, especially when handling images for authentication and beautification tasks.
Innovation Solution
The proposed solution involves an image processing method that utilizes a camera module and two processing units, one in a trusted execution environment (TEE) and another in a rich execution environment (REE), to acquire and process images. This method includes acquiring a target image, calculating depth information using speckle images, and performing face recognition, ensuring high safety and speed by switching between execution environments for different processing tasks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If image processing is performed in a single execution environment, then device complexity is reduced, but data safety and processing speed cannot be simultaneously ensured
Solution Approach 1:
The patent divides the execution environment into two separate segments: TEE (Trusted Execution Environment) for high-security operations like authentication, and REE (Rich Execution Environment) for general-purpose processing. This segmentation allows each environment to be optimized independently for its specific purpose, ensuring data safety in TEE while maintaining processing efficiency in REE, thereby resolving the contradiction between reliability and device complexity.
2Measurement precision
If high-precision processing is performed on all images, then measurement precision is improved, but loss of time increases due to processing overhead
Solution Approach 1:
The patent applies different processing qualities to different images based on their purpose: authentication images processed in TEE receive high-precision processing to ensure security, while other images processed in REE use standard processing. This local quality differentiation ensures measurement precision is improved only where necessary (authentication), thereby reducing overall processing time and resolving the contradiction between precision and time loss.
3Measurement precision
If image accuracy is increased for all processing tasks, then measurement precision is improved, but use of energy increases
Solution Approach 1:
The patent implements local quality by processing only authentication-critical images with high accuracy in the energy-optimized TEE environment, while other images are processed in REE with standard accuracy settings. This selective approach ensures measurement precision is improved only where it matters most (authentication), thereby reducing overall energy consumption and resolving the contradiction between image accuracy and energy use.
Data Source
AI summary
An image processing method, an image processing device, a computer readable storage medium and an electronic device are disclosed. The method includes: operating a camera module to acquire a target image in response to a first processing unit receiving an image acquisition instruction; and performing a predetermined processing on the target image.