Abstract Land cover change analysis was carried out using Moderate Resolution Imaging Spectroradiometer (MODIS) Enhanced Vegetation Index (EVI) time series data for the period 2005–2014. MODIS EVI data coupled with Quality Assessment Science Data Sets (QASDS) was de-noised with Savitzky–Golay filter while enhancing quality and preserving the temporal profile of EVI. Wavelet transform (WT) based approach along with Sen slope’s method was used for land cover change and trend analysis. The WT based approach is useful for studying multiscale and non-stationary processes. Mann–Kendall test was performed to confirm the significance of the identified trends. Proposed approach identified 358 locations as change points, where 285 (79.6%) and 73 (20.4%) locations were detected as ‘Change’ and ‘False Change’ with respect to high resolution images. The proposed approach is useful for monitoring land cover changes that provide vital inputs for sustainable management of land resources.

中文翻译：

---

使用时间序列 EVI 数据进行长期土地覆盖趋势分析的基于小波和非参数统计的方法

摘要 利用中分辨率成像光谱仪 (MODIS) 增强型植被指数 (EVI) 时间序列数据对 2005-2014 年期间的土地覆盖变化进行了分析。MODIS EVI 数据与质量评估科学数据集 (QASDS) 结合使用 Savitzky-Golay 滤波器进行降噪，同时提高质量并保留 EVI 的时间分布。基于小波变换 (WT) 的方法以及 Sen 坡度的方法用于土地覆盖变化和趋势分析。基于 WT 的方法对于研究多尺度和非平稳过程很有用。进行 Mann-Kendall 检验以确认已识别趋势的显着性。提议的方法将 358 个位置确定为变化点，其中 285 个（79.6%）和 73 个（20.4%）位置被检测为高分辨率图像的“变化”和“假变化”。