Update app.py

app.py

@@ -7,11 +7,11 @@ import pytesseract
 # Initialize EasyOCR reader
 easyocr_reader = easyocr.Reader(['en'])
 
-# Explicitly set Tesseract command path for Hugging Face Spaces
+# Explicitly set Tesseract binary path (Hugging Face Spaces default)
 pytesseract.pytesseract.tesseract_cmd = '/usr/bin/tesseract'
 
 def get_metrics(model):
-    # Hardcoded metrics for demonstration
+    # Simulated metrics for demo
     if model == "EasyOCR":
         return {"Accuracy": "95%", "Precision": "94%", "Pipeline": "Easy Integration (90%)"}
     elif model == "Tesseract":
@@ -22,9 +22,9 @@ def get_metrics(model):
 def ocr_all(img, ground_truth):
     if img is None:
         return "No image provided", "No image provided"
-    
     img_array = np.array(img)
     
+    # EasyOCR
     try:
         easyocr_result = easyocr_reader.readtext(img_array, detail=0, paragraph=True)
         text_easy = "\n".join(easyocr_result)
@@ -32,25 +32,26 @@ def ocr_all(img, ground_truth):
         text_easy = f"EasyOCR Error: {str(e)}"
     metrics_easy = get_metrics("EasyOCR")
     
+    # Tesseract OCR with graceful error handling
     try:
         text_tess = pytesseract.image_to_string(img)
     except Exception as e:
         text_tess = f"Tesseract Error: {str(e)}"
     metrics_tess = get_metrics("Tesseract")
     
-    output_easy = f"**Output:**\n{text_easy}\n\n" + "\n".join([f"{k}: {v}" for k, v in metrics_easy.items()])
-    output_tess = f"**Output:**\n{text_tess}\n\n" + "\n".join([f"{k}: {v}" for k, v in metrics_tess.items()])
-
+    output_easy = f"**Output:**\n{text_easy}\n\n" + "\n".join([f"{k}: {v}" for k,v in metrics_easy.items()])
+    output_tess = f"**Output:**\n{text_tess}\n\n" + "\n".join([f"{k}: {v}" for k,v in metrics_tess.items()])
+    
     return output_easy, output_tess
 
 with gr.Blocks() as demo:
-    gr.Markdown("# OCR Demo on Hugging Face Spaces\nUpload an image and paste optional ground truth text.")
+    gr.Markdown("# OCR Demo with EasyOCR and Tesseract")
     img_in = gr.Image(type="pil", label="Upload Screenshot")
-    txt_in = gr.Textbox(lines=3, label="Paste Ground Truth Text")
+    txt_in = gr.Textbox(lines=4, label="Paste Ground Truth Text Here")
     btn = gr.Button("Run OCR")
     out_easy = gr.Markdown(label="EasyOCR Output")
     out_tess = gr.Markdown(label="Tesseract Output")
-    
+
     btn.click(ocr_all, inputs=[img_in, txt_in], outputs=[out_easy, out_tess])
 
 demo.launch()