Memory‐Processing‐Display Integrated Hardware with All‐In‐One Structure for Intelligent Image Processing

Author:

Shan Liuting12,Yu Rengjian12,Chen Zhenjia12,Zhang Xianghong12,Gao Changsong12,Lian Minrui12,Hu Yuanyuan3,Guo Tailiang12,Chen Huipeng12ORCID

Affiliation:

1. Institute of Optoelectronic Display National & Local United Engineering Lab of Flat Panel Display Technology Fuzhou University Fuzhou 350002 P. R. China

2. Fujian Science & Technology Innovation Laboratory for Optoelectronic Information of China Fuzhou 350100 P. R. China

3. Changsha Semiconductor Technology and Application Innovation Research Institute College of Semiconductors (College of Integrated Circuits) Hunan University Changsha 410082 P. R. China

Abstract

AbstractEmpowering displays with intelligent functions enables their application in intelligent image processing and intelligent interactive displays, which is the combination of future display and artificial intelligence technologies. However, existing display technologies face significant processing and transmission burdens due to their conventional hardware separation architecture, which separates memory from the processor and display module from the processing module. To address these challenges, a highly integrated memory‐ processing‐display intelligent light‐emitting hardware (MPDIH) capable of information generation, memory, processing, and visualization is proposed. The MPDIH, based on an organic light‐sensitive layer and activated by ultraviolet light irradiation, exhibits unique photoinhibition behavior attributed to the reverse light‐induced electric field formed by the directional arrangement of photo‐generated excitons. This behavior enables dynamic regulation and autonomous learning during the device training process. Leveraging this phenomenon, intelligent image processing is successfully demonstrated, achieving a significant contrast improvement with a maximum enhancement of 261% compared to unprocessed raw signals. Furthermore, the memory‐processing‐display intelligent image processing scheme enables fashion MNIST image recognition using an artificial neural network, and achieves noticeably higher recognition accuracy (>89%) compared to the original fuzzy images (<59%). Consequently, the proposed memory‐processing‐display intelligent light‐emitting hardware holds promise for future applications in integrated photonic systems and interactive artificial intelligence displays.

Funder

National Natural Science Foundation of China

Publisher

Wiley

Cited by 2 articles. 订阅此论文施引文献 订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3