update imagej

README.md

@@ -9,44 +9,13 @@ Intensity-based registration algorithm repository: https://github.com/uw-loci/sh
 |<img src="https://github.com/uw-loci/he_shg_synth_workflow/blob/master/thumbnails/he.jpg" width="320">|<img src="https://github.com/uw-loci/he_shg_synth_workflow/blob/master/thumbnails/shg.jpg" width="320">|
 
 ## Required packages
-Install required packages in a virtual environment, commands for anaconda/miniconda (https://www.anaconda.com/distribution/) are listed
-* python==3.6.x
+Install [anaconda/miniconda](https://docs.conda.io/en/latest/miniconda.html)  
+Required packages
 ```
-  conda create --name [NAME_ENV] python=3.6
-  conda activate [NAME_ENV]
+  $ conda env create --name syn --file env.yml
+  $ conda activate syn
 ```
-* matplotlib==3.1.2 
-```
-  conda install -c conda-forge matplotlib=3.1.2
-```  
-* numpy==1.17.4
-```
-  conda install -c anaconda numpy=1.17.4
-```  
-* pandas==0.25.3
-```
-  conda install -c anaconda pandas=0.25.3
-``` 
-* Pillow==5.3.0
-```
-  conda install -c anaconda pillow=5.3.0
-```  
-* pyimagej==0.4.0
-```
-conda install -c conda-forge pyimagej
-```
-* scikit-image==0.16.2
-```
-  conda install -c anaconda scikit-image=0.16.2
-```  
-* tqdm==4.42.0
-```
-  conda install -c conda-forge tqdm=4.42.0
-```  
-* pytorch>=1.3.1
-```
-  conda install pytorch torchvision cudatoolkit=10.1 -c pytorch
-```  
+
 ## Download example testing data, trained model weights, FIJI
 Execute download.py
 ```  

main.py

@@ -69,7 +69,7 @@ def main():
     parser.add_argument('--load-multiple-gpu-weights', type=int, default=1, help='1: multiple gpu weights, 0: single gpu weghts')
     parser.add_argument('--input-folder', type=str, default='default', help='input_test + _FOLDERNAME')
     parser.add_argument('--intensity', type=tuple, default=(20, 180), help='output intensity rescale')
-    parser.add_argument('--pilot', type=int, default=0, help='1: only process the first image, 0: process all images')
+    parser.add_argument('--pilot', type=int, default=1, help='1: only process the first image, 0: process all images')
 
     args = parser.parse_args()
 
@@ -104,7 +104,7 @@ def demo(args):
     model.to(device)
 
     print('loading ImageJ, please wait')
-    ij = imagej.init('fiji/fiji/Fiji.app/')
+    ij = imagej.init('fiji/Fiji.app/')
 
     # use for SHG
     TASK = args.input_folder
@@ -165,25 +165,27 @@ def demo(args):
                         io.imsave(OUTPUT_PATCH_DIR+img_name, result_patch)
 
         print('stitching, please wait...')                
+        os.makedirs(CHANNEL_DIR, exist_ok=True)
         params = {'type': 'Positions from file', 'order': 'Defined by TileConfiguration', 
                 'directory':OUTPUT_PATCH_DIR, 'ayout_file': 'TileConfiguration.txt', 
                 'fusion_method': 'Linear Blending', 'regression_threshold': '0.30', 
                 'max/avg_displacement_threshold':'2.50', 'absolute_displacement_threshold': '3.50', 
                 'compute_overlap':False, 'computation_parameters': 'Save computation time (but use more RAM)', 
                 'image_output': 'Write to disk', 'output_directory': CHANNEL_DIR}
         plugin = "Grid/Collection stitching"
-        ij.py.run_plugin(plugin, params)     
+        ij.py.run_plugin(plugin, params)    
 
         output_name = os.path.join(OUTPUT_DIR, fn)
         listOfChannels = [f for f in os.listdir(CHANNEL_DIR)]
         c1 = io.imread(os.path.join(CHANNEL_DIR, listOfChannels[0]))
         c1 = c1[:img.shape[0], :img.shape[1]]
         c1 = img_as_ubyte(c1)
         c1 = exposure.rescale_intensity(c1, in_range=args.intensity, out_range=(0, 255))
+        c1 = exposure.rescale_intensity(c1, in_range=(0, 255), out_range=(0, 1))
         print(str(k+1)+"/" + str(len(files)) + " output saved as: " + output_name)
         io.imsave(output_name, img_as_ubyte(c1))
         if args.pilot:
             break
 
 if __name__ == '__main__':
-    main()
+    main()