Merge pull request #3 from tinabe/nanoseg-demo

updates to the segmentation demo.

Author: dnjohnstone

06 Nanocrystal segmentation in SPED data - Demonstration on partly overlapping MgO cubes.ipynb

@@ -83,7 +83,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "dp = pxm.load_hspy('MgO2_16_TX5_c_bin2,2.hdf5',\n",
+    "dp = pxm.load_hspy('SPED_data_of_MgO_nanoparticles_bin2,2,1,1.hdf5',\n",
     "                  lazy=False,\n",
     "                  assign_to='electron_diffraction2d')"
    ]
@@ -167,7 +167,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "dp.align2D(shifts=shifts, crop=False, fill_value=0)"
+    "dp.align2D(shifts=shifts, crop=True)"
    ]
   },
   {
@@ -183,8 +183,8 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "scale = 0.03246\n",
-    "scale_real = 2.56\n",
+    "scale = 0.03246 \n",
+    "scale_real = 3.03\n",
     "dp.set_diffraction_calibration(scale)\n",
     "dp.set_scan_calibration(scale_real)\n",
     "\n",
@@ -284,7 +284,7 @@
    "cell_type": "code",
    "execution_count": null,
    "metadata": {
-    "scrolled": true
+    "scrolled": false
    },
    "outputs": [],
    "source": [
@@ -392,12 +392,9 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "gmags = unique_peaks.get_magnitudes()\n",
-    "gmags.data[gmags.data<10*scale] = 0\n",
-    "Gs = unique_peaks.data[np.where(gmags)]\n",
-    "Gs = pxm.DiffractionVectors(Gs)\n",
-    "print(np.shape(Gs)[0], ' unique vectors.')\n",
-    "Gs.axes_manager.set_signal_dimension(0)"
+    "Gs = unique_peaks.filter_vector_magnitudes(\n",
+    "    min_magnitude=10*scale, max_magnitude=np.inf)\n",
+    "print(np.shape(Gs)[0], ' unique vectors.')"
    ]
   },
   {
@@ -427,12 +424,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Optionally save and load the unique peaks\n",
-    "\n",
-    "`np.save('peaks.npy', Gs.data)\n",
-    "Gs = np.load('peaks.npy', allow_pickle=True)\n",
-    "Gs = pxm.DiffractionVectors(Gs)\n",
-    "Gs.axes_manager.set_signal_dimension(0)`"
+    "Optionally save and load the unique peaks"
    ]
   },
   {
@@ -523,7 +515,7 @@
    "source": [
     "min_distance = 5.5\n",
     "min_size = 10\n",
-    "max_size = None\n",
+    "max_size = np.inf\n",
     "max_number_of_grains = np.inf\n",
     "marker_radius = 2\n",
     "exclude_border = 2"
@@ -535,7 +527,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "i = 27\n",
+    "i = 26\n",
     "sep_i = separate_watershed(\n",
     "    VDFs.inav[i].data, min_distance=min_distance, min_size=min_size,\n",
     "    max_size=max_size, max_number_of_grains=max_number_of_grains,\n",
@@ -653,7 +645,9 @@
   {
    "cell_type": "code",
    "execution_count": null,
-   "metadata": {},
+   "metadata": {
+    "scrolled": true
+   },
    "outputs": [],
    "source": [
     "hs.plot.plot_images(corrsegs.segments, cmap='magma_r', axes_decor='off',\n",
@@ -688,7 +682,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### Create a signal mask for the direct beam\n",
     "Create a signal mask so that the region in the centre of each PED pattern, including the direct beam, can be excluded in the machine learning. "
    ]
   },
@@ -707,7 +700,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### Perform single value decomposition (SVD)"
+    "Perform single value decomposition (SVD)"
    ]
   },
   {
@@ -736,7 +729,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### Investigate the scree plot and use it as a guide to determine the number of components"
+    "Investigate the scree plot and use it as a guide to determine the number of components"
    ]
   },
   {
@@ -757,7 +750,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### NMF"
+    "NMF"
    ]
   },
   {
@@ -870,7 +863,6 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "#### Investigate the normalised cross-correlations\n",
     "Calculate the matrix of normalised cross-correlation for both the loadings and patterns first, to find suitable correlation threshold values. "
    ]
   },
@@ -890,7 +882,15 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "ncc_nmf = learn.get_ncc_matrix()\n",
+    "ncc_nmf = learn.get_ncc_matrix()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
     "ncc_nmf.plot(scalebar=False, cmap='RdBu')"
    ]
   },