Upload 2 files

app.py

@@ -0,0 +1,55 @@
+import gradio as gr
+from cellpose import models
+import numpy as np
+import cv2
+from PIL import Image
+
+# Load Cellpose pretrained model for cytoplasm
+model = models.Cellpose(model_type='cyto')  # or 'nuclei' for nucleus
+
+def segment_and_count(image: Image.Image):
+    # Convert image to numpy RGB
+    img_np = np.array(image.convert("RGB"))
+
+    # Run Cellpose
+    masks, flows, styles, diams = model.eval(img_np, diameter=None, channels=[0, 0])  # grayscale or RGB
+
+    # Create overlay image
+    outlines = masks_to_overlay(img_np, masks)
+
+    # Count unique masks (excluding background=0)
+    unique_ids = np.unique(masks)
+    cell_count = len(unique_ids[unique_ids > 0])
+
+    return Image.fromarray(outlines), f"Detected cells: {cell_count}"
+
+# Draw outlines on image
+def masks_to_overlay(image, masks):
+    overlay = image.copy()
+    contours, _ = cv2.findContours(masks.astype(np.uint8), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
+    cv2.drawContours(overlay, contours, -1, (0, 255, 0), 1)
+    return overlay
+
+# Gradio interface
+with gr.Blocks() as demo:
+    gr.Markdown("## 🧬 Cell Counting with Cellpose")
+    with gr.Row():
+        with gr.Column():
+            image_input = gr.Image(type="pil", label="Upload Microscopy Image")
+            run_btn = gr.Button("Count Cells")
+            count_output = gr.Textbox(label="Detected Cell Count")
+            examples = gr.Examples(
+                examples=[
+                    ["examples/cells1.png"],
+                    ["examples/cells2.jpg"],
+                    ["examples/cells3.png"]
+                ],
+                inputs=[image_input],
+                label="Example Images"
+            )
+        with gr.Column():
+            image_output = gr.Image(type="pil", label="Cell Mask Overlay")
+
+    run_btn.click(fn=segment_and_count, inputs=image_input, outputs=[image_output, count_output])
+
+demo.launch()

requirements.txt

@@ -0,0 +1,5 @@
+gradio
+cellpose
+numpy
+opencv-python
+Pillow