水稻在种植生产过程中容易受到气象因素的影响，如气温、光照日长和降雨量等。通过对影响水稻生长的温光序列数据进行分析，进而指导水稻的种植具有明显的实际意义。目前存在的问题有，一是当前度量温光序列的相似性方法没有考虑序列各个温光因子之间的权重问题，二是当前的提供水稻种植知识查询或指导的APP缺少对物候期内温光序列分析分功能。因此，本文设计并实现了一款基于Android的水稻物候期内温光序列分析预测工具。首先，为分析水稻物候期内的温光序列数据，本文实现了三种多元动态时间弯曲的相似性度量算法，分别是CPCA_SWDTW、HGSDTW、DTW，并使用KNN算法统计搜索分类准确率对算法进行了比较评价，得出了CPCA_SWDTW在K取值较大时准确率更高的结论。然后实现M-K算法对温光数据进行突变检验分析。最后，基于Android设计开发了水稻物候期内温光序列分析工具，提供用户管理，历史温光序列数据查询，水稻物候期知识查询，历史温光序列数据的相似性分析、差异性分析、突变检测，以及气温预测预警等主要功能。
目录
摘要 １
关键词 １
Abstract １
引言
引言
1绪论 ２
1.1研究背景 ２
1.2水稻物候期内温光序列分析工具研究进展 ２
1.3基于动态时间弯曲的时间序列相似性度量的研究进展 ２
1.4目前存在的问题 ３
1.5研究内容与研究路线 ３
1.5.1研究内容 ４
1.5.2技术路线 ４
1.6论文组织与结构 ５
2水稻物候期内温光序列分析工具的需求分析及开发相关知识 ５
2.1系统功能分析 ５
2.2系统用例图 ６
2.2数据库设计 ７
2.2.1系统ER图设计 ７
2.2.2数据库表设计 １１
2.3系统开发环境及相关技术 １３
2.3.1服务器端开发环境简介 １３
2.3.2客户端开发环境简介 １３
2.3.3服务器端和客户端的交互 １３
2.3.4使用到的相关技术和API接口 １４
3温光序列数 *51今日免费论文网|www.jxszl.com +Q: #351916072# 
据分析模块的设计与实现 １５
3.1相似性分析 １５
3.1.1功能分析 １５
3.1.2相似性度量算法的原理和实现 １６
3.1.3功能测试 ２１
3.2差异性分析 ２２
3.3突变检测 ２２
3.3.1功能分析 ２２
3.3.2 MK突变检测算法原理及实现 ２３
3.3.3功能测试 ２４
4查询和管理模块的设计与实现 ２５
4.1用户管理子模块 ２５
4.1.1功能分析 ２５
4.1.2用户注册 ２５
4.1.3用户登录 ２７
4.1.4用户修改个人信息 ２８
4.2历史温光数据查询子模块 ３１
4.2.1功能分析 ３１
4.2.2功能实现 ３３
4.2.3功能测试 ３３
4.3物候期知识查询子模块 ３４
4.3.1功能分析 ３５
4.3.2功能实现 ３５
4.3.3功能测试 ３５
5预测预警模块的设计与实现 ３６
5.1天气预报 ３６
5.1.1获取方法 ３６
5.1.2结果展示 ３７
5.2气温预测 ３７
6总结和展望 ３７
6.1总结 ３７
6.2展望 ３８
致谢 ３９
参考文献： ４０
水稻物候期内温光序列分析APP的设计与实现
Design and Implementation of App for Prediction of Temperature and Light Sequences in Rice Phenological Period
Student majoring in Computer Science and Technology HU Ling
Tutor JIANG Haiyan
Abstract：Rice is vulnerable to meteorological factors, such as temperature, sunshine and rainfall in the process of planting and production. In the era of big data, a large amount of meteorological data has been accumulated. It has obvious practical significance to guide rice cultivation by analyzing the temperaturelight time series that affect rice growth. The existing problems are as follows: firstly, the current methods of measuring the similarity of thermooptical sequence do not consider the weight of each thermooptical factor in the sequence; secondly, the current APPs which provide knowledge query or guidance for rice cultivation lack the function of subanalysis of thermooptical time series in phenological period. Therefore, this paper designs and implements a temperaturelight time series analysis and prediction tool based on Android for rice phenophase. Firstly, in order to analyze the temperaturelight time series in rice phenological period, three similarity measurement algorithms of multivariate dynamic time warping were implemented, namely CPCA_SWDTW, HGSDTW and DTW. The accuracy of the algorithm was compared and evaluated by using KNN algorithm, and a conclusion was drawn that CPCA_SWDTW has higher accuracy when K value is larger. Then the MK algorithm is implemented to test and analyze the mutation of temperaturelight time series. Finally, based on Android, I designed and developed a temperaturelight sequence analysis tool for rice phenological period, which provides user management, historical temperaturelight time series query, rice phenological knowledge query, similarity analysis, difference analysis, mutation detection, and temperature prediction and early warning.

原文链接：http://www.jxszl.com/jsj/jsjkxyjs/563930.html