Results 41 to 50 of about 6,061 (205)
Research and Application of Face Recognition Based on Deep Convolutional Neural Networks [PDF]
, 2016 人脸识别是计算机视觉领域中研究的热点之一，不仅是因为以自然脸为研究对象是非常有挑战性的，而且有很多应用需要将人脸识别作为第一步。近年来，随着智能手机的普及，每天有成千上万的照片被上传到社交网络平台，如微博、微信等等，其中包括大量的人脸照片。深度学习是近十年来人工智能领域取得的重要突破，深度学习通过建立类似于人脑的分层模型结构，对输入数据逐级提取从底层到高层的特征，从而能很好地建立从底层信号到高层语义的映射关系。本文以深度学习的方法在人脸识别中的研究与应用为研究对象。 论文首先介绍了研究的背景和意义 ...
万经勇
core
Deep learning based physical layer wireless communication techniques:opportunities and challenges [PDF]
, 2019 The development of the fifth-generation wireless communications (5G) system is promoted by the high requirements of the high reliability and super-high network capacity.However,existing communication techniques are hard to achieve the high requirements ...
Guan GUI, Hao HUANG, Yu WANG
core +1 more source
Rare Metals, EarlyView.ABSTRACT The high accuracy in surface‐enhanced Raman scattering‐lateral flow immunoassays (SERS–LFIAs) is critical for reliable point‐of‐care testing (POCT) in clinical diagnostics. Conventional approaches are often affected by sampling variability and uneven distribution of immunoprobes, leading to unreliable signal fluctuations.
Shuai Zhao +9 more
wiley +1 more source
智能科学与技术学报, 2020 针对常用的信号传输模型存在使用场景单一、预测精度不佳的问题，提出一种适用于多场景的数据驱动无线信号传输模型。首先根据先验知识从预处理后的数据构造初始特征，接着进行特征选择，以得到输入特征集合。然后分析建模需求，选择深度置信网络（DBN）、残差网络（ResNet）和堆叠自编码器（SAE）作为区间二型模糊规则的后件（个体深度学习器），经过区间二型模糊推理进行集成。最后采用5G网络信号传输实测数据，并进行实验验证。结果表明，3种个体深度学习器在测试集上的表现均优于Cost231-Hata模型和反向传播神经网络（
赵亮 +4 more
doaj
Conservation Biology, EarlyView.Abstract Consumer demand for ivory perpetuates the unsustainable and illegal killing of African elephants and other wildlife species. Interventions that aim to change consumer behavior are increasingly recognized as a crucial element of demand management. However, poor design and implementation have limited their effectiveness.
Molly R. C. Brown +2 more
wiley +1 more source
Research on Scene-specific Pedestrian Detection Method Based on Deep Learning [PDF]
, 2015 行人检测一直是计算机视觉中，尤其是目标检测领域的研究重点。而由于环境的时变性和多样性，不同场景之间样本不满足同分布，普通场景下训练得到的检测器直接应用到某一特定场景时性能会急剧下降。基于这样的背景，本文依据场景的复杂程度不同提出了两种对应的解决方法，主要工作和创新点如下： 1）当背景比较简单时，容易获取样本的标定信息，本文提出了一种快速的行人检测算法，首先基于结构化局部边缘模式计算样本特征，然后通过积分图技术快速计算分类器分数，从而使得检测过程更加高效 ...
袁德东
core
Research Progress on Automatic Assessment of Bone Age#br# [PDF]
, 2020 Bone age is an important indicator of human growth and development, which can objectively reflect the growth level and maturity of individuals. Traditional manual bone age assessment usually compares the X-ray of the left wrist with the reference ...
ZHAN Meng-jun, ZHANG Shi-jie, CHEN Hu,et al.
core +1 more source
From Clueless to Confident: How ChatGPT Transforms Academic Writing in Chinese as a Second Language
International Journal of Applied Linguistics, EarlyView.ABSTRACT Generative artificial intelligence (GenAI) has demonstrated its potential to enhance the quality of second‐language (L2) academic writing. This study used a qualitative approach comprising analysis of ChatGPT usage‐history screenshots, written assignments, and semi‐structured interview data on students’ use of GenAI in their L2 Chinese ...
Lanfang Sun +4 more
wiley +1 more source
中西医结合护理, 2023 Objective To evaluate the application of multiple interactive teaching methods in clinical nursing teaching and practice. Methods From August 2021 to April 2022, a total of 96 nursing students who participated in the internship in the hospital were ...
WANG Shangying (王尚莹) +4 more
doaj +1 more source
Research of Pedestrian Detection Based on Cascade Convolutional Neural Networks and Multiple-Cue Fusion [PDF]
, 2016 作为计算机视觉的一个重要研究方向，行人检测具有很高的研究意义和应用 价值。其可为目标跟踪、人体行为理解等研究领域以及智能驾驶系统、安全监控 系统、高级人机交互系统等应用系统的设计提供基础性技术支持。然而，由于人 体外形多变，这使得行人检测易受人体动作、背景、遮挡和光照等因素的影响。 如何克服这些因素所带来的影响是当前计算机视觉研究的热点和难点。 本文首先介绍行人检测系统的基本概念和经典技术，总结近二十年来该领域 的发展情况，并对深度学习技术在计算机视觉领域的应用予以分析和整理，然后 ...
冯麒峻
core