[Advances in deep learning multimodal fusion for early diagnosis of knee osteoarthritis]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za ZhiTian J, Ma J, Zhao J, Ma X.
europepmc +1 more source
[Construction and validation of a multimodal risk prediction model for burn sepsis in infants and toddlers based on deep learning]. [PDF]
Zhonghua Shao Shang Yu Chuang Mian Xiu Fu Za ZhiShen YH, Zhang L, Ji Y, Chen S, Shen WM.
europepmc +1 more source
[Multi-modal tumor boundary clustering recognition based on artificial intelligence virtual cells and near-infrared surgical field]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za ZhiLi W, Wang T, Yang X, Zhou F.
europepmc +1 more source
[Deep learning models based on fused ultrasound images for the assessment of cystocele in women]. [PDF]
Zhong Nan Da Xue Xue Bao Yi Xue BanRan S, Lu R, Li M, Qu C.
europepmc +1 more source
A New Research on Image Fusion
, 2007 该文首先概述了数据融合的基本概念和应用,概括了现有图像融合的基本原理和进行融合的基本步骤,介绍了三种不同层次的图像融合方法.通过Matlab仿真程序设计,对传统的图像融合方法和现有的融合方法进行了比较和评价,分析了图像融合领域存在的问题,最后提出了基于Contourlet和并行PCNN算法的图像融合方法,并预测了未来的融合研究方向.Summarized the basic concept of data fusion and its application,introduced the steps of
闫敬文, 卓琳, 屈小波
core
[An auxiliary diagnosis system for cervical intraepithelial neoplasia based on colposcopic images]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za ZhiGu X +6 more
europepmc +1 more source
[Synergistic learnable frequency and multi-scale spatial network for lightweight skin cancer classification]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za ZhiZhuang J, Lyu Z, Li X.
europepmc +1 more source
[Application of hybrid attention guided dual-path residual learning in magnetic resonance imaging diagnosis]. [PDF]
Sheng Wu Yi Xue Gong Cheng Xue Za ZhiZhu W, Cai X, Liu Y.
europepmc +1 more source
, 2011 目的:为了获取多光谱图像数据,实现对多光谱CCD相机图像 信号进行显示、存储和回放,并以此为依据对相机的功能和成像质量进行检测,我们设计了一套多光谱CCD相机专用的图像采集系统。方法:系统硬件采用 CPLD与FIFO相结合的方案对相机输出的数字图像数据进行整合和缓存,结合基于PCI总线的Camlink图像采集卡,使用Streams5图像采集 软件,最终实现了多光谱图像数据的采集。结果:该系统四个通道可以同时同步采集四个谱段的多光谱数据,每个谱段图像宽度为3072个像素,10bit位 宽。结论 ...
张伯珩 +5 more
core
Gaoya dianqi电力无人机巡检图像中复杂的背景、异物的遮挡以及多种天气下的光照影响使得绝缘子的准确识别仍是一个具有挑战性的任务。同时，现有的绝缘子检测算法不能有效表征多姿态和大长宽比的绝缘子串目标且算法参数量过大。因此，文中提出一种融合多方位注意力信息聚合与网络剪枝的链式绝缘子检测方法来检测复杂背景下的绝缘子目标。首先，该方法采用点—链式结构有效表征了多姿态的绝缘子串目标；其次，网络模型使用Drn-c42骨干网络从无人机巡检图像中提取绝缘子串图像特征；然后，通过多方位注意力聚合网络对图像进行特征细化并使网络模型以局部 ...
汤永恒, 孙水发, 李旭琛
doaj