QGIS and Applications in Agriculture and Forest

<p>Introduction xi</p>
<p>Chapter 1. Coupling Radar and Optical Data for Soil Moisture Retrieval over Agricultural Areas&nbsp; 1<br />Mohammad EL HAJJ, Nicolas BAGHDADI, Mehrez ZRIBI and Hassan BAZZI</p>
<p>1.1. Context 1</p>
<p>1.2. Study site and satellite data &nbsp;2</p>
<p>1.2.1. Radar images 2</p>
<p>1.2.2. Optical image 4</p>
<p>1.2.3. Land cover map 4</p>
<p>1.3. Methodology 5</p>
<p>1.3.1. Inversion approach of radar signal for estimating soil moisture 5</p>
<p>1.3.2. Segmentation of crop and grasslands areas 6</p>
<p>1.3.3. Soil moisture mapping &nbsp;8</p>
<p>1.4. Implementation of the application via QGIS . 10</p>
<p>1.4.1. Layout 10</p>
<p>1.4.2. Radar images 14</p>
<p>1.4.3. Optical image 20</p>
<p>1.4.4. Land cover map 26</p>
<p>1.4.5. Segmentation of crop s areas and grasslands 26</p>
<p>1.4.6. Elimination of small spatial units &nbsp;29</p>
<p>1.4.7. Mapping soil moisture &nbsp;33</p>
<p>1.4.8. Soil moisture maps &nbsp;43</p>
<p>1.5. Bibliography 44</p>
<p>Chapter 2. Disaggregation of Thermal Images 47<br />Mar BISQUERT and Juan Manuel S&Aacute;NCHEZ</p>
<p>2.1. Definition and context &nbsp;47</p>
<p>2.2. Disaggregation method &nbsp;48</p>
<p>2.2.1. Image pre–processing &nbsp;48</p>
<p>2.2.2. Disaggregation 50</p>
<p>2.3. Practical application of the disaggregation method . 53</p>
<p>2.3.1. Input data 53</p>
<p>2.3.2. Step 1: pre–processing &nbsp;54</p>
<p>2.3.3. Step 2: disaggregation &nbsp;63</p>
<p>2.4. Results analysis 73</p>
<p>2.5. Bibliography 75</p>
<p>Chapter 3. Automatic Extraction of Agricultural Parcels from Remote Sensing Images and the RPG Database with QGIS/OTB 77<br />Jean–Marc GILLIOT, Camille LE PRIOL, Emmanuelle VAUDOUR and Philippe MARTIN</p>
<p>3.1. Context 77</p>
<p>3.2. Method of AP extraction &nbsp;79</p>
<p>3.2.1. Formatting the RPG data &nbsp;79</p>
<p>3.2.2. Classification of SPOT satellite images &nbsp;81</p>
<p>3.2.3. Intersect overlay between extracted AP and FB with crop validation 81</p>
<p>3.3. Practical application of the AP extraction &nbsp;82</p>
<p>3.3.1. Software and data 83</p>
<p>3.3.2. Setting up the Python script &nbsp;86</p>
<p>3.3.3. Step 1: formatting the RPG data &nbsp;89</p>
<p>3.3.4. Step 2: classification of SPOT satellite Images . 97</p>
<p>3.3.5. Step 3: intersect overlay between extracted AP and FB and crop validation 110</p>
<p>3.4. Acknowledgements 116</p>
<p>3.5. Bibliography 116</p>
<p>Chapter 4. Land Cover Mapping Using Sentinel–2 Images and the Semi–Automatic Classification Plugin: A Northern Burkina Faso Case Study 119<br />Louise LEROUX, Luca CONGEDO, Beatriz BELL&Oacute;N, Raffaele GAETANO and Agn&egrave;s B&Eacute;GU&Eacute;</p>
<p>4.1. Context 119</p>
<p>4.2. Workflow for land cover mapping &nbsp;120</p>
<p>4.2.1. Introduction to SCP and S2 images &nbsp;120</p>
<p>4.2.2. Pre–processing 122</p>
<p>4.2.3. Land cover classification &nbsp;126</p>
<p>4.2.4. Classification accuracy assessment and post–processing 129</p>
<p>4.3. Implementation with QGIS and the plugin SCP 131</p>
<p>4.3.1. Software and data 131</p>
<p>4.3.2. Step 1: data pre–processing &nbsp;133</p>
<p>4.3.3. Step 2: land cover classification &nbsp;139</p>
<p>4.3.4. Step 3: assessment of the classification accuracy and post–processing 144</p>
<p>4.4. Bibliography 150</p>
<p>Chapter 5. Detection and Mapping of Clear–Cuts with Optical Satellite Images &nbsp;153<br />Kenji OSE</p>
<p>5.1. Definition and context &nbsp;153</p>
<p>5.2. Clear–cuts detection method &nbsp;154</p>
<p>5.2.1. Step 1: change detection geometric and radiometric pre–processing 154</p>
<p>5.2.2. Steps 2 and 3: forest delimitation &nbsp;160</p>
<p>5.2.3. Step 4: clear–cuts classification &nbsp;160</p>
<p>5.2.4. Steps 5 and 6: export in vector mode &nbsp;162</p>
<p>5.2.5. Step 7: statistical evaluation. &nbsp;164</p>
<p>5.2.6. Method limits 166</p>
<p>5.3. Practical application 166</p>
<p>5.3.1. Software and data 166</p>
<p>5.3.2. Step 1: creation of the changes image &nbsp;168</p>
<p>5.3.3. Steps 2 and 3: creation, merging and integration of masks 170</p>
<p>5.3.4. Step 4: clear–cuts detection &nbsp;174</p>
<p>5.3.5. Step 5: vector conversion &nbsp;177</p>
<p>5.4. Bibliography 180</p>
<p>Chapter 6. Vegetation Cartography from Sentinel–1 Radar Images &nbsp;181<br />Pierre–Louis FRISON and C&eacute;dric LARDEUX</p>
<p>6.1. Definition and context &nbsp;181</p>
<p>6.2. Classification of remote sensing images &nbsp;183</p>
<p>6.3. Sentinel–1 data processing &nbsp;185</p>
<p>6.3.1. Radiometric calibration &nbsp;186</p>
<p>6.3.2. Ortho–rectification of calibrated data &nbsp;186</p>
<p>6.3.3. Clip over a common area &nbsp;187</p>
<p>6.3.4. Filtering to reduce the speckle effect &nbsp;187</p>
<p>6.3.5. Generation of color compositions based on different polarizations 188</p>
<p>6.4. Implementation of the processing within QGIS 189</p>
<p>6.4.1. Downloading data &nbsp;194</p>
<p>6.4.2. Calibration, ortho–rectification and stacking of Sentinel–1 data over a common area &nbsp;198</p>
<p>6.4.3. Speckle filtering 201</p>
<p>6.4.4. Other tools 202</p>
<p>6.5. Data classification 205</p>
<p>6.6. Bibliography 212</p>
<p>Chapter 7. Remote Sensing of Distinctive Vegetation in Guiana Amazonian Park &nbsp;215<br />Nicolas KARASIAK and Pauline PERBET</p>
<p>7.1. Context and definition &nbsp;215</p>
<p>7.1.1. Global context 215</p>
<p>7.1.2. Species 216</p>
<p>7.1.3. Remote sensing images available &nbsp;217</p>
<p>7.1.4. Software 219</p>
<p>7.1.5. Method implementation &nbsp;219</p>
<p>7.2. Software installation 220</p>
<p>7.2.1. Dependencies installation available in OsGeo . 220</p>
<p>7.2.2. Installation of scikit–learn &nbsp;221</p>
<p>7.2.3. Dzetsaka installation &nbsp;222</p>
<p>7.3. Method 222</p>
<p>7.3.1. Image processing 223</p>
<p>7.3.2. Cloud mask creation &nbsp;225</p>
<p>7.4. Processing 227</p>
<p>7.4.1. Creating training plots &nbsp;227</p>
<p>7.4.2. Classification with dzetsaka plugin &nbsp;230</p>
<p>7.4.3. Post–classification &nbsp;236</p>
<p>7.5. Final processing 239</p>
<p>7.5.1. Synthesis of predicted images &nbsp;240</p>
<p>7.5.2. Global synthesis and cleaning unwanted areas . 242</p>
<p>7.5.3. Statistical validation limits &nbsp;244</p>
<p>7.6. Conclusion 245</p>
<p>7.7. Bibliography 245</p>
<p>Chapter 8. Physiognomic Map of Natural Vegetation 247<br />Samuel ALLEAUME and Sylvio LAVENTURE</p>
<p>8.1. Context 247</p>
<p>8.2. Method 247</p>
<p>8.2.1. Segmentation of the VHSR mono–date image . 249</p>
<p>8.2.2. Calculation of temporal variability indices 249</p>
<p>8.2.3. Extraction of natural vegetation using time series 251</p>
<p>8.2.4. Vegetation densities &nbsp;252</p>
<p>8.2.5. Maximum productivity index of herbaceous areas 255</p>
<p>8.3. Implementation of the application &nbsp;256</p>
<p>8.3.1. Study area 256</p>
<p>8.3.2. Software and data 257</p>
<p>8.3.3. Step 1: VHSR image processing &nbsp;259</p>
<p>8.3.4. Step 2: calculation of the variability indices on the time series 264</p>
<p>8.3.5. Step 3: extraction of the natural vegetations from the time series of Sentinel–2 image by thresholding method 267</p>
<p>8.3.6. Step 4: classification of vegetation density by supervised classification SVM 274</p>
<p>8.3.7. Step 5: extraction of the level of productivity of grasslands 277</p>
<p>8.3.8. Step 6: final map 279</p>
<p>8.4. Bibliography 282</p>
<p>Chapter 9. Object–Based Classification for Mountainous Vegetation Physiognomy Mapping &nbsp;283<br />Vincent THIERION and Marc LANG</p>
<p>9.1. Definition and context &nbsp;283</p>
<p>9.2. Method for detecting montane vegetation physiognomy 284</p>
<p>9.2.1. Satellite image pre–processing &nbsp;286</p>
<p>9.2.2. Image segmentation &nbsp;289</p>
<p>9.2.3. Sampling, learning and segmented image classification 291</p>
<p>9.2.4. Statistical validation of classification &nbsp;295</p>
<p>9.2.5. Limits of the method &nbsp;297</p>
<p>9.3. Application in QGIS 298</p>
<p>9.3.1 Pre–processing 299</p>
<p>9.3.2. Segmentation 312</p>
<p>9.3.3. Classification 319</p>
<p>9.4. Bibliography 337</p>
<p>List of Authors 341</p>
<p>Index 343</p>
<p>Scientific Committee 347</p>