Update app.py

app.py

@@ -4,23 +4,19 @@ import numpy as np
 from fpdf import FPDF
 import tempfile
 import os
-from PIL import Image
 from paddleocr import PaddleOCR
 import time
 
-# Initialize PaddleOCR with updated parameters
+# Initialize PaddleOCR once with updated parameters
 ocr_model = PaddleOCR(use_textline_orientation=True, lang='en')
 
 def analyze_uv_coverage(img, brightness_threshold=150):
     gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
-    
-    # Apply threshold to identify sterilized vs unsterilized zones
     _, binary_mask = cv2.threshold(gray, brightness_threshold, 255, cv2.THRESH_BINARY)
 
-    # Apply morphological operations for better segmentation
     kernel = np.ones((5, 5), np.uint8)
-    binary_mask = cv2.dilate(binary_mask, kernel, iterations=1)  # Dilation to fill gaps
-    
+    binary_mask = cv2.dilate(binary_mask, kernel, iterations=1)
+
     total_pixels = binary_mask.size
     sterilized_pixels = cv2.countNonZero(binary_mask)
     coverage_percent = (sterilized_pixels / total_pixels) * 100
@@ -35,8 +31,8 @@ def analyze_uv_coverage(img, brightness_threshold=150):
 def create_pdf_report(coverage_percent, extracted_texts, annotated_image_path, output_path):
     pdf = FPDF()
     pdf.add_page()
-    
-    pdf.set_font("Arial", 'B', size=16)
+
+    pdf.set_font("Arial", 'B', 16)
     pdf.cell(200, 10, txt="UV Sterilization Report", ln=True, align='C')
     pdf.ln(10)
 
@@ -53,7 +49,6 @@ def create_pdf_report(coverage_percent, extracted_texts, annotated_image_path, o
         pdf.cell(0, 8, "No text detected.", ln=True)
 
     pdf.ln(10)
-    
     pdf.cell(0, 10, "Annotated Image:", ln=True)
     pdf.image(annotated_image_path, x=10, y=pdf.get_y(), w=pdf.w - 20)
 
@@ -62,13 +57,20 @@ def create_pdf_report(coverage_percent, extracted_texts, annotated_image_path, o
 def process_image(input_img, brightness_threshold=150):
     img = cv2.cvtColor(np.array(input_img), cv2.COLOR_RGB2BGR)
 
+    # Resize large images for faster processing
+    max_dim = 640
+    h, w = img.shape[:2]
+    if max(h, w) > max_dim:
+        scale = max_dim / max(h, w)
+        img = cv2.resize(img, (int(w * scale), int(h * scale)))
+
     start_time = time.time()
-    ocr_result = ocr_model.ocr(np.array(input_img))
+    ocr_result = ocr_model.ocr(img)
     ocr_time = time.time() - start_time
 
     extracted_texts = []
     for line in ocr_result:
-        if line:  # Skip None or empty entries
+        if line:
             for word_info in line:
                 extracted_texts.append(word_info[1][0])
 
@@ -83,9 +85,8 @@ def process_image(input_img, brightness_threshold=150):
 
     annotated_img_rgb = cv2.cvtColor(annotated_img, cv2.COLOR_BGR2RGB)
 
-    # Hide extracted texts in output; show only coverage percentage
     report_text = f"UV Sterilization Coverage: {coverage_percent:.2f}%"
-    # Optionally include OCR processing time
+    # Optionally include OCR time:
     # report_text += f"\n\nOCR Processing Time: {ocr_time:.2f} seconds"
 
     os.unlink(temp_img_file.name)
@@ -107,5 +108,7 @@ iface = gr.Interface(
     description="Upload a post-UV sterilization image to analyze surface coverage and generate a compliance report."
 )
 
+iface.queue()  # Enable request queuing to improve UX on heavy processing
+
 if __name__ == "__main__":
     iface.launch()