Iris Recognition Using Low-Resolution Quality Inspection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional iris recognition systems face delays in processing high-resolution images, leading to non-real-time recognition and increased power consumption due to lengthy data processing and image quality inspection times, especially when inspecting images based on low-resolution images.
Innovation Solution
An electronic device with an image sensor, image processing unit, and iris recognition unit that processes images in parallel by creating a low-resolution second image for quality inspection, allowing real-time iris recognition by using a high-resolution image stored in a buffer only after the low-resolution image meets predetermined quality conditions, while providing user guidance for proper image capture.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If image quality inspection is performed on high-resolution images, then measurement precision is improved, but processing time increases and real-time recognition cannot be achieved
Solution Approach 1:
The patent divides the image processing task into two segments: low-resolution image quality inspection and high-resolution iris recognition. The image quality inspection is performed on a low-resolution version of the image, which reduces the data volume and processing time while still providing sufficient information to assess image quality. This segmentation allows the system to quickly determine whether an image is suitable for iris recognition without processing the entire high-resolution image for quality inspection purposes.
Solution Approach 2:
The patent creates a low-resolution copy of the original high-resolution image for the purpose of image quality inspection. This copy contains sufficient visual information to evaluate image quality metrics such as focus, brightness, and composition, but requires significantly less computational resources to process. The original high-resolution image is preserved for the actual iris recognition process, ensuring that the full image quality is available when needed.
2Measurement precision
If high-resolution images are processed for iris recognition, then recognition accuracy is improved, but power consumption increases
Solution Approach 1:
The processing workload is segmented into two stages: first, a low-resolution image is processed for quality inspection to determine suitability for iris recognition; second, only if the quality check passes, is the high-resolution image processed for iris recognition. This segmentation avoids the unnecessary power consumption of processing high-resolution images that fail the quality check, while still maintaining high recognition accuracy for qualifying images.
Solution Approach 2:
The patent performs a preliminary action (low-resolution image quality inspection) before the main action (high-resolution iris recognition). This preliminary step filters out unsuitable images early in the process, preventing the system from wasting power processing high-resolution images that would not yield successful recognition results. The preliminary action consumes minimal power but significantly reduces overall power consumption by avoiding unnecessary high-resolution processing.
3Productivity
If low-resolution images are used for quality inspection, then processing speed is improved, but recognition delay occurs due to time gap between image capture and inspection
Solution Approach 1:
The system performs preliminary action by continuously generating and processing low-resolution images for quality inspection in advance. This allows the quality assessment to be completed before the user needs recognition, eliminating the perceived delay. The low-resolution image processing occurs in the background while the user interacts with the device, so when recognition is requested, the quality check has already been performed or is in progress.
Solution Approach 2:
The patent maintains continuous useful action by continuously capturing and inspecting low-resolution images in real-time as the user interacts with the device. This continuous inspection process ensures that when the user initiates iris recognition, the system already has recent quality assessment data available, minimizing any delay between image capture and recognition. The continuous processing keeps the inspection pipeline constantly active rather than batch-processing images.
4Productivity
If parallel processing of high-resolution and low-resolution images is implemented, then real-time recognition is achieved, but device complexity increases
Solution Approach 1:
The patent segments the processing pipeline into distinct functional blocks: a low-resolution image generation unit, a quality inspection unit operating on low-resolution images, and a high-resolution image processing unit for iris recognition. Each segment handles a specific task, allowing independent optimization and simplifying the overall system design. The segmentation enables parallel processing where the low-resolution inspection occurs simultaneously with high-resolution image capture, achieving real-time performance without requiring a completely complex rearchitecture of the entire system.
Data Source
AI summary
An electronic device, according to various embodiments of the present disclosure, may include: an image sensor that obtains the first image; an image processing unit that processes the obtained first image to thereby create the second image; an image quality inspecting unit that performs image quality inspection on the basis of the created second image; and an iris recognition unit that, if the image quality of the second image satisfies a predetermined condition, performs iris recognition on the basis of the obtained first image.