基于多帧聚合的电容屏超分辨率成像方法  

A Capacitive Screen Super-Resolution Imaging Method Based on Multi-Frame Aggregation

作　　者：朱晓静[1] 乔晓晓 刘芹 ZHU Xiaojing;QIAO Xiaoxiao;LIU Qin(Key Laboratory of Aerospace Information Security and Trusted Computing,Ministry of Education,School of Cyber Science and Engineering,Wuhan University,Wuhan 430072,Hubei,China)

机构地区：[1]空天信息安全与可信计算教育部重点实验室,武汉大学国家网络安全学院,湖北武汉430072

出　　处：《武汉大学学报(理学版)》2024年第6期715-724,共10页Journal of Wuhan University:Natural Science Edition

基　　金：国家自然科学基金(62272348)。

摘　　要：随着智能终端的推广,电容式触摸屏(以下简称“电容屏”)得到了广泛普及,但受限于硬件设计和成像原理,目前电容屏的成像分辨率较低,该应用的进一步发展受到了限制。为了应对这一挑战,通过采集导电物体运动过程中的电容图像序列,利用基于多帧聚合的电容屏超分辨率方法对该图像序列进行处理,实现高分辨率电容图像的重建。该超分辨率方法包含三个关键阶段:预处理、图像重建和视觉优化。在预处理阶段,通过YOLOv8n进行目标检测并使用Nano跟踪器进行跟踪,去除了背景噪声,得到了干净的低分辨率电容图像序列。随后,通过图像重建步骤对每帧图像进行Lanczos上采样、对齐和合并,最大程度地保持图像细节的完整性和保真性。最后,在视觉优化阶段提出了电容图像去模糊算法,经由实验评估,验证了所提出算法具有最佳去模糊效果,能够有效提升图像的清晰度和边缘细节。为了验证该超分辨率方法的有效性和必要性,在大量数据集上进行了实验。实验结果显示,该方法显著提高了电容图像的分辨率和视觉质量。With the widespread adoption of smart devices,capacitive touchscreens(“capacitive screens”)have become increasingly popular.However,further development is constrained by its relatively low imaging resolution due to the hardware design and imaging principles.To overcome this limitation,we propose a super-resolution imaging method that reconstructs high-resolution capacitive images by aggregating multiple frames captured during the movement of conductive objects.The proposed super-resolution algorithm comprises three essentials procedures:preprocessing,image reconstruction,and visual optimization.In the preprocessing stage,YOLOv8n and Nano tracker are employed respectively for object detection and object tracking,enabling background noise removal,thus resulting in clean low-resolution capacitive image sequences.During the image reconstruction phase,each frame undergoes Lanczos upsampling,alignment,and merging,ensuring the preservation of image detail and fidelity.Finally,the visual optimization stage introduces a capacitive image deblurring algorithm,which,as validated by experiments,exhibits the best deblurring effect,effectively enhancing image clarity and edge details.Extensive experiments are conducted on a large dataset to evaluate the effectiveness and necessity of the super-resolution algorithm proposed in this paper.The experimental results demonstrate significant advantages of the proposed algorithm in terms of imaging performance,and effectively improve the image resolution and visual quality of capacitive images.

关 键 词：电容式触摸屏 超分辨率方法 电容图像 视觉优化 

分 类 号：TP309.2[自动化与计算机技术—计算机系统结构]