Results 11 to 20 of about 380 (97)
An Intelligent Method for Pork Freshness Identification Based on EfficientNet Model [PDF]
, 2023 A method for measuring pork freshness based on images and the EfficientNet framework was established. A total of 2 500 images of pork with different freshness were collected as original dataset and processed by image enhancement to construct a new ...
LIU Chao, ZHANG Jiayu, QI Chao, HUANG Jichao, CHEN Kunjie
core +1 more source
智能科学与技术学报, 2022 以机器人为代表的自主智能体系统在工业、服务业等众多领域具有广泛的应用前景和重要的应用价值，但现有自主智能体系统在运动灵巧性、感知信息的完备性、复杂任务和环境的适应性等方面都面临巨大的挑战，难以在开放环境下像人一样灵巧、精准地完成各种复杂操作。而随着以深度学习、深度强化学习为代表的人工智能技术的成功，以操作技能学习为核心的理论研究成为突破自主智能体精准灵巧操作的重要方向，在汲取、融汇联结主义和符号主义的思想精华后构建的操作技能学习框架有望取代传统的示教编程、手工编程模式，以自主学习 ...
王硕
doaj +1 more source
, 2019 针对濒死鱼难以实时、高精度检测的问题,提出一种基于迁移学习与卷积神经网络的濒死鱼预警系统。通过迁移学习方法优化卷积神经网络的初始权重,使模型在训练初期即具有一定的泛化能力,优化前网络总体误差为1.75,优化后为0.75。为进一步提升模型性能,使用两组卷积神经网络作为特征提取器进行对比,结果显示,ZFnet网络(AP=0.902)略好于VGGnet网络(0.898),两者均显著优于常见的HOG+SVM算法(AP=0.259)。江苏省农业重大新品种创制项目(PZCZ201742 ...
周辉, 尹绍武, 李宏志, 谢万里
core +1 more source
Xibei Gongye Daxue Xuebao, 2023 In view of the problem that the generalization ability of traditional data-driven fault diagnosis model declines or even fails in mechanical system diagnosis, a fault diagnosis method based on multi-scale transfer symbolic dynamic entropy and support ...
YU Guangwei, YAN Li
doaj +1 more source
Weakly supervised semantic segmentation and optimization algorithm based on multi-scale feature model [PDF]
, 2019 In order to improve the accuracy of weakly-supervised semantic segmentation method,a segmentation and optimization algorithm that combines multi-scale feature was proposed.The new algorithm firstly constructs a multi-scale feature model based on transfer
Changzhen XIONG, Hui ZHI
core +1 more source
Dizhen xuebao2023年中国建成了全球规模最大的地震预警网，预警网中的台站配置了速度计、加速度计、简易烈度计（MEMS）等多种地震仪器。目前用于深度学习模型训练的数据集主要由速度计数据组成，不同类型仪器记录的数据在处理时精度会下降。为了解决这一问题，本文使用Yu等（2023）发布的BRNN模型作为预训练模型；然后使用2万6 739条人工标注的四川地区的速度计、加速度计和MEMS数据进行迁移学习。实验结果表明：迁移学习显著提升了模型对加速度计和MEMS预警仪器数据记录到的Pg，Sg震相的检测精度 ...
Yuqi Cai +6 more
doaj +1 more source
Gaoya dianqi, 2022 气体绝缘金属封闭开关设备的局部放电具有随机性，即使同一类型缺陷仍有较多表观变化形式，传统模式识别构造的特征难以适应其多变性；近年来兴起的卷积神经网络虽具有自适应提取特征的能力，但训练得到性能更好的网络模型一方面需要增加网络深度，另一方面需要更多的数据样本作为支撑。鉴于此，提出了基于在ImageNet数据集上训练的VGG、InceptionV3、Resnet50 3种网络模型迁移学习的GIS局部放电模式识别方法，并将网络提取的特征应用于在小数据集下表现良好的经典分类器SVM ...
唐志国, 曹智, 何宁辉
doaj
, 2019 针对自然生态蝴蝶多种特征检测的实际需求,以及生态环境下蝴蝶检测效率低、精度差问题,本文提出了一种基于迁移学习和可变形卷积深度神经网络的蝴蝶检测算法(Transfer learning and deformable convolution deep learning network,TDDNET).该算法首先使用可变形卷积模型重建ResNet-101卷积层,强化特征提取网络对蝴蝶特征的学习,并以此结合区域建议网络(Region proposal network, RPN)构建二分类蝴蝶检测网络 ...
张栋 +5 more
core +2 more sources
Video description method based on multidimensional and multimodal information [PDF]
, 2020 In order to solve the problem of complex information representation in automatic video description tasks,a multi-dimensional and multi-modal visual feature extraction and fusion method was proposed.Firstly,multi-dimensional features such as static and ...
Enjie DING +3 more
core +1 more source
Research Progress on Computer-Aided Skeleton-Based Individual Identification in Forensic Radiology [PDF]
, 2021 Individual identification based on imaging data of the skeleton of a corpse is a key technique for forensic identification. To reduce the influence of artificial factors, computer-aided semi-automatic or automatic individual identification has become one
LI Yuan, ZHAO Huan, LIANG Wei-bo, et al.
core +1 more source