面向对象的特色农作物种植遥感调查方法研究
作者简介:单治彬(1993-),男,陕西安康人,硕士生,主要从事遥感反演研究。E-mail: zbshan_chd@126.com
收稿日期: 2018-02-28
要求修回日期: 2018-07-19
网络出版日期: 2018-10-17
基金资助
宁夏回族自治区重点研发计划项目(2016KJHM130)
Remote Sensing Investigation Method of Object-oriented Crops with Special Characteristics
Received date: 2018-02-28
Request revised date: 2018-07-19
Online published: 2018-10-17
Supported by
Key Research and Development Project of the Ningxia Hui Autonomous Region, No.2016KJHM130.
Copyright
宁夏自治区具有土地、光能、引黄灌溉等优势,为宁夏特色农作物(硒砂瓜、枸杞、大枣)的生长提供了先天条件。快速准确地获取特色农作物的种植信息不仅是宁夏特色农作物监测、估产和灾害评估的重要依据,同时也是分析特色农作物结构分布变化和评价区域特色农业生产影响的重要凭证。近年来,随着航天技术和卫星传感器的不断发展,越来越多的学者将遥感技术运用到农作物种植信息的提取研究中。但是传统的遥感调查模型都是基于中低分辨遥感数据建立的,对于新的高分数据没有完备的信息提取模型。此外,基于GF-1遥感影像对类似宁夏特色农作物(硒砂瓜、枸杞、大枣)的信息提取研究相对较少,决策条件和分类模型的选择也难以满足高分农业的需求。基于此,本文利用国产GF-1 PMS遥感影像,在分析3类特色农作物光谱特征和纹理特征的基础上,建立了面向对象的支持向量机(SVM)分类模型,总体分类精度达到94.94%,Kappa系数为0.9174。同时将分类结果与传统的SVM分类结果相比较,研究发现面向对象的SVM模型的精度更高,效果最好,纹理信息的引入使光谱特征差异较小的枸杞和大枣更容易区分,有效降低了模型错分和漏分误差,改善了模型分类结果。研究结果为实现宁夏特色农作物的快速自动化提取提供了有效途径,也为开展农作物承保和受灾定损评估体系建设提供技术支撑。
单治彬 , 孔金玲 , 张永庭 , 李欢 , 关红 , 胡永新 , 李健锋 , 张文博 . 面向对象的特色农作物种植遥感调查方法研究[J]. 地球信息科学学报, 2018 , 20(10) : 1509 -1519 . DOI: 10.12082/dqxxkx.2018.180116
The advantages of Ningxia Hui Autonomous Region is land, light energy, and irrigation of Yellow River, which provides an inherent condition for the growth of characteristic crops of Ningxia (e.g. watermelon, Chinese wolfberry and jujube). Accessing to planting structure information of characteristic crops quicklf and accurately is not only an important basis for regional crop monitoring, yield estimation and disaster assessment, but also an important evidence for analyzing the spatial pattern changes of characteristic crop and assessing the impact of regional characteristics on agricultural production. In recent years, with the continuous development of space technology and remote sensing satellites, more and more scholars have applied remote sensing technology to the extraction of crop planting structure information. However, the traditional remote sensing survey model is only applicable to low resolution and medium resolution remote sensing data, and domestic and abroad scholars have relatively few studies on information extraction of similar Ningxia characteristic crops (e.g. watermelon, Chinese wolfberry and jujube), and the selection of classification models and strategies is difficult to meet the demand of rapid monitoring, accurate acquisition and real-time decision-making. Based on this, this paper calculate and analyze the spectral characteristics and texture features of the three types of specialty crops under the support of GF-2 remote sensing data, and establish an SVM of object-oriented classification model, the overall classification accuracy is 94.94% and the Kappa coefficient is 0.9174. And compare the classification results with the traditional SVM classification results, The study found that the SVM of object-oriented model established in this paper has the highest accuracy and best results, the texture information makes it easier to distinguish the Chinese wolfberry and jujube,whice has theless difference in spectral characteristics, and the texture information effectively reducing the model error and missing error, and improving the model classification results.
Fig. 1 Study area location diagram图1 研究区位置示意图(NIR-Red-Green假彩色合成) |
Tab. 1 Statistics of the ground quadrat data表1 地面样方数据统计 |
作物类型 | 训练数据 | 检验数据 | |||||
---|---|---|---|---|---|---|---|
样方数 | 样方总面积/km2 | 面积比/% | 样方数 | 样方总面积/km2 | 面积比/% | ||
硒砂瓜 | 100 | 83.56 | 39.43 | 70 | 53.31 | 39.47 | |
枸杞 | 100 | 69.38 | 32.74 | 70 | 46.95 | 34.76 | |
大枣 | 60 | 58.99 | 27.83 | 40 | 34.82 | 25.77 |
Fig. 2 Characteristic curve of crop characteristic图2 特色农作物特征曲线 |
Fig. 3 Characteristic curve of crop texture feature图3 特色农作物纹理特征曲线 |
Fig. 4 Texture feature results of different windows图4 不同窗口的纹理特征结果 |
Fig. 5 Different segmentation combined threshold results图5 不同分割合并阈值结果 |
Fig. 6 Remote sensing classification map of special crop planting structure图6 特色农作物种植结构遥感分类图 |
Tab. 2 Confusion matrix results表2 混淆矩阵结果 |
地物类型 | 硒砂瓜 | 枸杞 | 大枣 | 合计 |
---|---|---|---|---|
未分类 | 0 | 0 | 0 | 0 |
硒砂瓜 | 39 557 | 361 | 0 | 39 918 |
枸杞 | 439 | 13 190 | 116 | 13 745 |
大枣 | 0 | 2986 | 20 492 | 23 478 |
合计 | 39 996 | 16 537 | 20 608 | 77 141 |
Tab. 3 Accuracy evaluation result表3 精度评定结果 |
地物类型 | 硒砂瓜 | 枸杞 | 大枣 | |
---|---|---|---|---|
制图精度/% | 98.90 | 79.76 | 99.44 | |
用户精度/% | 99.10 | 95.96 | 87.28 | |
错分误差/% | 0.90 | 4.04 | 12.72 | |
漏分精度/% | 1.10 | 20.24 | 0.56 | |
总体精度/% | 94.94 | |||
Kappa系数 | 0.917 |
Tab. 4 Accuracy comparison results of different classification methods表4 不同分类方法精度对比表 |
类别 | 光谱单数据源SVM法 | 光谱结合纹理SVM法 | 面向对象的SVM法 | |||||
---|---|---|---|---|---|---|---|---|
制图精度/% | 用户精度/% | 制图精度/% | 用户精度/% | 制图精度/% | 用户精度/% | |||
硒砂瓜 | 90.02 | 95.84 | 92.3 | 99.58 | 98.90 | 99.10 | ||
枸杞 | 76.30 | 66.72 | 78.52 | 72.84 | 79.76 | 95.96 | ||
大枣 | 80.78 | 76.94 | 91.24 | 80.74 | 99.44 | 87.28 | ||
总体精度/% | 84.82 | 89.17 | 94.94 | |||||
Kappa系数 | 0.769 | 0.815 | 0.917 |
The authors have declared that no competing interests exist.
[1] |
[
|
[2] |
|
[3] |
|
[4] |
|
[5] |
[
|
[6] |
|
[7] |
[
|
[8] |
[
|
[9] |
[
|
[10] |
[
|
[11] |
[
|
[12] |
|
[13] |
[
|
[14] |
[
|
[15] |
[
|
[16] |
|
[17] |
[
|
[18] |
[
|
[19] |
[
|
[20] |
[
|
[21] |
[
|
[22] |
|
/
〈 |
|
〉 |