8.1.2

Author: semiautomaticgit

__init__.py

@@ -34,7 +34,7 @@ def description():
 
 
 def version():
-    return 'Version 8.1.2 - Infinity'
+    return 'Version 8.1.3 - Infinity'
 
 
 def icon():

docs/repository.xml

@@ -1,9 +1,9 @@
 <?xml version = '1.0' encoding = 'UTF-8'?>
 <plugins>
-    <pyqgis_plugin name="Semi-Automatic Classification Plugin - master" version="8.1.1" plugin_id="284">
+    <pyqgis_plugin name="Semi-Automatic Classification Plugin - master" version="8.1.3" plugin_id="284">
         <description><![CDATA[The Semi-Automatic Classification Plugin (SCP) allows for the supervised classification of remote sensing images, providing tools for the download, the preprocessing and postprocessing of images.]]></description>
         <about><![CDATA[Developed by Luca Congedo, the Semi-Automatic Classification Plugin (SCP) allows for the supervised classification of remote sensing images, providing tools for the download, the preprocessing and postprocessing of images. Search and download is available for Landsat, Sentinel-2 images. Several algorithms are available for the land cover classification. This plugin requires the installation of Remotior Sensus, GDAL, OGR, Numpy, SciPy, and Matplotlib. For more information please visit https://fromgistors.blogspot.com .]]></about>
-        <version>8.1.2</version>
+        <version>8.1.3</version>
         <qgis_minimum_version>3.0.0</qgis_minimum_version>
         <qgis_maximum_version>3.99.0</qgis_maximum_version>
         <homepage><![CDATA[https://fromgistors.blogspot.com/p/semi-automatic-classification-plugin.html]]></homepage>

interface/classification_tab.py

@@ -150,12 +150,16 @@ def save_classifier_action():
 
 
 # perform classification
-def run_classifier(save_classifier=None, preview_point=None):
+def run_classifier(
+        save_classifier=None, preview_point=None,
+        classification_confidence=None
+):
     threshold = False
     signature_raster = False
     cross_validation = True
     find_best_estimator = False
-    classification_confidence = False
+    if classification_confidence is None:
+        classification_confidence = False
     input_normalization = load_classifier = class_weight = None
     rf_max_features = rf_number_trees = rf_min_samples_split = svm_c = None
     svm_gamma = svm_kernel = mlp_hidden_layer_sizes = None
@@ -490,33 +494,47 @@ def run_classifier(save_classifier=None, preview_point=None):
                 cfg.util_qgis.save_qml_style(
                     raster, '%s/%s.qml' % (directory, name)
                 )
-            if output.extra['algorithm_raster'] is not None:
-                # add raster to layers
-                cfg.util_qgis.add_raster_layer(
-                    output.extra['algorithm_raster']
-                )
-            if output.extra['signature_rasters'] is not None:
+            if 'algorithm_raster' in output.extra:
+                if output.extra['algorithm_raster'] is not None:
+                    # add raster to layers
+                    cfg.util_qgis.add_raster_layer(
+                        output.extra['algorithm_raster']
+                    )
+            if 'signature_rasters' in output.extra:
                 # add raster to layers
-                for s in output.extra['signature_rasters']:
-                    cfg.util_qgis.add_raster_layer(s)
+                try:
+                    for s in output.extra['signature_rasters']:
+                        if s is not None:
+                            cfg.util_qgis.add_raster_layer(s)
+                except Exception as err:
+                    str(err)
     else:
         cfg.mx.msg_err_1()
     cfg.ui_utils.remove_progress_bar(smtp=str(__name__))
     return output
 
 
 # create classification preview
-def create_preview(preview_point):
+def create_preview(preview_point, classification_confidence=None):
     point = cfg.utils.check_point_in_image(point=preview_point)
     if point is False:
         cfg.mx.msg_war_3()
         return False
     cfg.preview_point = point
-    output = run_classifier(preview_point=point)
+    output = run_classifier(
+        preview_point=point,
+        classification_confidence=classification_confidence
+    )
     if output is None:
         return False
     elif output.check:
-        output_raster = output.path
+        if classification_confidence is None:
+            output_raster = output.path
+        else:
+            if 'algorithm_raster' in output.extra:
+                output_raster = output.extra['algorithm_raster']
+            else:
+                output_raster = None
         # move previous preview to group
         group = cfg.util_qgis.group_index(
             cfg.qgis_registry[cfg.reg_group_name]
@@ -539,7 +557,8 @@ def create_preview(preview_point):
         else:
             macroclass = False
         # apply symbology
-        apply_class_symbology(cfg.classification_preview, macroclass)
+        if classification_confidence is None:
+            apply_class_symbology(cfg.classification_preview, macroclass)
         # move to top
         cfg.util_qgis.move_layer_to_top(cfg.classification_preview)
         cfg.util_qgis.set_group_visible(group, False)
@@ -567,7 +586,7 @@ def pointer_left_click(point):
 
 # right click pointer
 def pointer_right_click(point):
-    create_preview(point)
+    create_preview(point, classification_confidence=True)
 
 
 # set script button

metadata.txt

@@ -2,7 +2,7 @@
 name=Semi-Automatic Classification Plugin
 qgisMinimumVersion=3.00
 description=The Semi-Automatic Classification Plugin (SCP) allows for the supervised classification of remote sensing images, providing tools for the download, the preprocessing and postprocessing of images.
-version=8.1.2
+version=8.1.3
 about=Developed by Luca Congedo, the Semi-Automatic Classification Plugin (SCP) allows for the supervised classification of remote sensing images, providing tools for the download, the preprocessing and postprocessing of images. Search and download is available for Landsat, Sentinel-2 images. Several algorithms are available for the land cover classification. This plugin requires the installation of Remotior Sensus, GDAL, OGR, Numpy, SciPy, and Matplotlib. For more information please visit https://fromgistors.blogspot.com .
 
 author=Luca Congedo
@@ -13,6 +13,9 @@ icon=semiautomaticclassificationplugin.png
 
 changelog=
 
+	8.1.3
+	-fixed classification tool
+
 	8.1.2
 	-fixed classification tool