Quality-Weighted Multimodal Image Matching for User Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The accuracy of target identification is affected by varying image quality due to complex photographing environments in offline use, as different acquisition modes yield images of differing quality, impacting matching and identification processes.
Innovation Solution
A method to determine a user account by obtaining multiple types of images using different acquisition modes, assigning weighting factors based on image quality, and using these factors to calculate identification results, ensuring accurate matching and identification by prioritizing high-quality images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple types of images are acquired using different acquisition modes, then the identification accuracy can be improved through weighted evaluation, but the device complexity increases due to multiple image acquisition systems
Solution Approach 1:
The patent segments the image acquisition system into multiple independent acquisition modes (color image acquisition, infrared image acquisition, depth image acquisition), each using dedicated sensors and processing pipelines. This segmentation allows each subsystem to be optimized independently while their results are combined through weighted evaluation to achieve high identification accuracy without requiring a completely complex integrated system.
Solution Approach 2:
The patent implements a universal image processing framework that handles multiple types of images (color, infrared, depth) through a common weighted evaluation mechanism. The processing circuitry is designed to universally process different image types by determining weighting factors based on image quality metrics and combining them through a unified identification algorithm, reducing the need for separate processing paths for each image type.
2Measurement precision
If weighting factors are determined based on image quality in complex photographing environments, then the identification result accuracy is improved, but the processing time and computational complexity increase
Solution Approach 1:
The patent performs preliminary evaluation of image quality metrics for each acquired image type before the main identification process. By determining the quality of color, infrared, and depth images in advance and calculating their weighting factors beforehand, the system prepares the weighted evaluation parameters ahead of time, reducing the computational burden during the actual identification process and minimizing processing time delays.
Solution Approach 2:
The patent dynamically adjusts weighting factors based on image quality parameters such as clarity, brightness, and signal-to-noise ratio. When image quality is good, higher weights are assigned to that image type; when quality degrades due to complex photographing conditions, the system automatically reduces its weight and relies more on other image types, thereby maintaining identification accuracy while adapting to varying processing requirements.
3Reliability
If multiple types of images are processed with different weighting factors, then the impact of high-quality images is enhanced and low-quality images are reduced, but the computational complexity of the processing circuitry increases
Solution Approach 1:
The patent applies local quality assessment to each image type independently, evaluating specific quality metrics (such as clarity for color images, thermal contrast for infrared images, and depth accuracy for depth images) separately. This allows the system to determine weighting factors based on the local quality characteristics of each image type rather than requiring a complex global analysis of all images simultaneously, reducing processing circuitry complexity while maintaining reliable identification.
Data Source
AI summary
In a method for determining a user account, a plurality of images of a to-be-identified target is obtained, including a plurality of types of images. Each type of the plurality of types of images is obtained using a different acquisition mode. For each type of image, a weighting factor is determined based on an image quality. For each candidate user account, an identification result is obtained based on the weighting factors and image matching degrees of the plurality of types of images. The image matching degree indicates a degree of feature matching between the to-be-identified target and the respective candidate user account based on the respective type of image. Based on the identification results, the user account that matches the to-be-identified target is determined from the plurality of candidate user accounts. Apparatus and non-transitory computer-readable storage medium counterpart embodiments are also contemplated.