排序方式: 共有38条查询结果,搜索用时 24 毫秒
1.
目的 卫星遥感技术具有覆盖范围广、探测周期短、调查成本低等优势而广泛应用于大区域农作物分类。然而在种植结构复杂区(如城乡结合部),因其地块破碎、同期生长的作物种类多且分布分散,利用传统的统计分类或机器学习方法进行农作物分类时仍存在精度不高的问题。为提高种植结构复杂区农作物分类精度。方法 文章选取河北省廊坊市广阳区为研究区,以GF-1 PMS全色多光谱融合影像为数据源,采用U-Net、PSPNet及DeepLabv3+,3种深度学习模型进行农作物分类研究。分析模型参数对农作物分类精度的影响,评价3种深度学习模型的农作物分类精度,优选农作物精细分类方法。结果 (1)学习率与3种深度学习模型的分类精度呈正相关关系,较大的学习率(0.01,0.001)下,3种模型收敛速度快,分类精度高。批样本量与模型分类稳定性相关,批样本量设为100时,3种模型的分类稳定性最好。(2)相比PSPNet、DeepLabv3+模型,U-Net模型分类效果最好,总体分类精度为89.32%。(3)GF-1 PMS影像结合U-Net模型可有效提升种植结构复杂区农作物分类精度,大宗作物春玉米、夏玉米的分类精度在80%以上,花生、红薯、蔬菜小宗作物分类精度在60%以上。结论 该研究可为准确获取种植结构复杂区的农作物类型、面积及空间分布信息提供参考依据。 相似文献
2.
《Journal of Teaching in Travel & Tourism》2013,13(4):33-51
Abstract Experiential learning approaches such as role-play have been found to be valuable methods of bridging the divide between academic knowledge and practical skills, a problem often cited in tourism and hospitality management education. Such approaches have been found to contribute towards deeper learning by enhancing students' interest, motivation, participation, knowledge and skill development. This paper reports on the implementation of an experiential learning approach designed to encourage and facilitate deeper learning approaches, with the contributing aims of providing students with a more interesting learning experience and a broader set of skills for future employment. 相似文献
3.
针对辐射源目标精确识别需求,结合以深度学习为代表的机器学习理论技术,提出将改进型AlexNet作为特征提取器,实现目标细微特征提取固化,形成智能化识别网络模型。以广播式自动相关监视(ADS-B)信号为实验对象,在机场实地采集了13个目标的ADS-B脉冲信号数据作为辐射源目标个体识别的训练和测试样本,利用AlexNet和改进的AlexNet验证了算法的有效性。结果表明,改进的AlexNet网络训练时间更快,综合识别率达到98.32%. 相似文献
4.
Richard A. Miller 《The Journal of economic education》2013,44(2):142-150
Professor Miller describes a game which has been used in intermediate price theory courses as well as in introductory classes. Two versions are presented—a duopoly and a six-firm market, both with undifferentiated product and a known market demand schedule. Students are also asked to make a monopoly of the six firms by forming a trust or holding company. Miller contends that the game is more stimulating than lectures aud suspects that it will result in longer retention. 相似文献
5.
Len Hand Peter Sanderson Mike O'Neil 《Accounting Education: An International Journal》2013,22(2):103-111
Pressures on the unit of resources within higher education are noted and the possible implications for assessment practice considered. There is a tension between low-cost and high-quality assessments, yet assessment appears to drive the students' agenda and requires careful thought if the programme design is to lead to high-quality learning outcomes. A theoretical framework is presented with the purpose of developing an assessment strategy which will enhance the quality of learning by the student. A case study of an introductory accounting module is described and issues of teaching, learning and assessment are explored. A message of hope is signalled; it may be possible to foster deep learning through assessments which engage the student actively with the subject, despite the apparent constraints acting upon academics. 相似文献
6.
车辆类型识别方法是智能交通系统的关键技术之一。利用深度学习的高维特征泛化学习能力,将改进的LeNet-5卷积神经网络用于基于交通微波雷达的大小车型分类识别。首先,以雷达触发前的N帧信号为基础,对雷达的回波信号进行分析并构建数据集;然后,分析LeNet-5卷积神经网络的特点;最后提出一种改进的LeNet-5卷积神经网络。实验结果表明,与传统的支持向量机方法相比,所提方法能够智能学习大小车的雷达时频信号特征,大小车型识别准确率达到97%以上,可为交通场景下的车型识别研究提供新的技术途径。 相似文献
7.
针对跳频信号分选存在人工提取参数特征具有复杂性的问题,提出了一种基于深度学习的识别方法。首先对跳频信号进行短时傅里叶变换,得到二维的时频矩阵;接着提取信号的轮廓特征,构造三维矩阵作等高线图,并对等高线图进行预处理;最后把预处理后的等高线图输入到卷积神经网络中进行训练、测试,进而实现分类识别。仿真结果表明,在不需要复杂的人工提取参数特征的基础上,在分选率为100〖WT《Times New Roman》〗%〖WTBZ〗时,所提方法经裁剪处理下的信噪比为-15 dB,比支持向量机和传统K-Means聚类算法都低10 dB。实测数据的算法验证表明,所提方法能够将大疆精灵4Pro、hm无人机、司马航模X8HW以及大疆悟2这四类无人机正确分类。 相似文献
8.
社交物联网是社交网络概念在物联网中整合后兴起的一个蓬勃发展的研究领域。提出了一种适用于社交物联网网络的改进型节点级信任模型,并通过与其他信任模型的对比仿真实验证明在恶意节点的攻击下,提出的模型拥有更好的稳定性和适用性,总体波动较小。同时,针对实际社交物联网网络中新加入网络的陌生节点可能遇到的网络延迟影响信任值评估的问题,在改进型节点级信任模型的基础上进一步使用了深度学习模型对其进行信任值预测。仿真证明,使用深度学习预测后模型的系统性能明显优于不使用深度学习的模型,成功交互率提升约1.8%。 相似文献
9.
针对高移动用户在毫米波系统中用户与基站之间频繁切换增大延迟开销问题,提出了一种新颖的集成深度学习协调波束成形(Deep Learning Coordinated Beamforming,DLCBF)解决方案。该方案通过协调多个基站同时为一个移动用户服务,接收用户上行导频序列和预编码码本训练模型,进而预测最佳下行波束向量,克服了大型天线阵列毫米波系统中选取最佳波束成形向量需要的巨大训练开销。仿真结果表明,所提方案与传统的毫米波波束成形策略相比有更高的频谱有效率。 相似文献
10.