tile FtsZ rescale for bacteria

Author: wl-stepp

NNfeeder.py

@@ -12,7 +12,7 @@
 from SmartMicro.ImageTiles import getTilePositionsV2
 
 
-def prepareNNImages(mitoFull, drpFull, model):
+def prepareNNImages(mitoFull, drpFull, model, bacteria = False):
     """Preprocess raw iSIM images before running them throught the neural network.
 
     Args:
@@ -74,13 +74,16 @@ def prepareNNImages(mitoFull, drpFull, model):
         i = 0
         inputData = np.zeros((positions['n']**2, nnImageSize, nnImageSize, 2), dtype=np.uint8())
         for position in positions['px']:
+
             inputData[i, :, :, :] = inputDataFull[:,
                                                   position[0]:position[2],
                                                   position[1]:position[3],
                                                   :]
-            # inputData[i, :, :, 1] =  exposure.rescale_intensity(
-            #    inputData[i, :, :, 1], (0, np.max(inputData[i, :, :, 1])),
-            #    out_range=(0, 255))
+            if bacteria:
+                inputData[i, :, :, 1] =  exposure.rescale_intensity(
+                    inputData[i, :, :, 1], (0, np.max(inputData[i, :, :, 1])),
+                    out_range=(0, 255))
+
             inputData[i, :, :, 0] = exposure.rescale_intensity(
                 inputData[i, :, :, 0], (0, np.max(inputData[i, :, :, 0])),
                 out_range=(0, 255))

NetworkWatchdog.py

@@ -60,6 +60,7 @@ class NetworkWatchdog(QWidget):
     def __init__(self):
         # Set if channels are used in Micromanager
         self.channels = True
+        self.bacteria = True
 
         # Read settings from the json file depending on which computer we are on
         with open('./ATS_settings.json') as file:
@@ -264,7 +265,7 @@ def onModified(self, event):
 
         # Preprocess the data and make tiles if necessary
         inputData, positions = prepareNNImages(
-            mitoFull, drpFull, self.model)
+            mitoFull, drpFull, self.model, self.bacteria)
         # print(inputData.shape)
         # Calculate the prediction on the full batch of images
         outputPredict = self.model.predict_on_batch(inputData)