Add full logging and preprocessed image visualization

app.py

@@ -2,6 +2,8 @@ import gradio as gr
 import cv2
 import numpy as np
 from model import TrafficSignDetector
+import sys
+import io
 
 # Load the detector
 detector = TrafficSignDetector('config.yaml')
@@ -10,8 +12,9 @@ def detect_traffic_signs(image):
     """
     Process the uploaded image and return the image with detected signs.
     :param image: PIL Image or numpy array
-    :return: numpy array of processed image
+    :return: tuple of (detected image, preprocessed image)
     """
+    # Redirect stdout to capture all logs
     print(f"Received image type: {type(image)}")
     # Convert PIL to numpy if necessary
     if hasattr(image, 'convert'):
@@ -22,13 +25,14 @@ def detect_traffic_signs(image):
     image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
     print(f"Converted to BGR, shape: {image.shape}")
 
-    # Perform detection
-    result_image = detector.detect(image)
+    # Perform detection (returns tuple of (detected_image, preprocessed_image))
+    result_image, preprocessed_image = detector.detect(image)
 
     # Convert back to RGB for Gradio
     result_image = cv2.cvtColor(result_image, cv2.COLOR_BGR2RGB)
+    preprocessed_image = cv2.cvtColor(preprocessed_image, cv2.COLOR_BGR2RGB)
 
-    return result_image
+    return result_image, preprocessed_image
 
 # Create Gradio interface
 with gr.Blocks(title="Traffic Sign Detector") as demo:
@@ -39,8 +43,16 @@ with gr.Blocks(title="Traffic Sign Detector") as demo:
         input_image = gr.Image(label="Upload Image", type="pil")
         output_image = gr.Image(label="Detected Signs")
     
+    with gr.Row():
+        preprocessed_image = gr.Image(label="Preprocessed Image (640x640, Letterboxed)")
+    
     detect_btn = gr.Button("Detect Traffic Signs")
-    detect_btn.click(fn=detect_traffic_signs, inputs=input_image, outputs=output_image)
+    detect_btn.click(
+        fn=detect_traffic_signs, 
+        inputs=input_image, 
+        outputs=[output_image, preprocessed_image],
+        queue=True  # Enable queue to ensure logs are shown
+    )
 
 if __name__ == "__main__":
-    demo.launch(server_name="0.0.0.0", server_port=7860)
+    demo.queue().launch(server_name="0.0.0.0", server_port=7860)

model.py

@@ -151,9 +151,10 @@ class TrafficSignDetector:
         """
         Perform inference on the image and draw bounding boxes.
         :param image: numpy array of the image
-        :return: image with drawn bounding boxes
+        :return: tuple of (image with drawn bounding boxes, preprocessed image for visualization)
         """
-        print(f"Input image shape: {image.shape}")
+        print(f"\n{'='*80}")
+        print(f"INPUT IMAGE shape: {image.shape}")
         
         # Store original image for drawing (uint8)
         original_image = image.copy()
@@ -165,6 +166,9 @@ class TrafficSignDetector:
         # Normalize pixel values for inference
         image = self._preprocess(image)
         
+        # Store preprocessed image for visualization (convert back to 0-255 for display)
+        preprocessed_display = (image * 255).astype(np.uint8) if image.max() <= 1.0 else image.copy()
+        
         # Use imgsz=640 to match training size
         # Use iou_threshold for NMS (Non-Maximum Suppression) to remove overlapping boxes
         results = self.model(image, conf=self.conf_threshold, imgsz=640, iou=0.45)
@@ -207,4 +211,5 @@ class TrafficSignDetector:
                     label = f"{self.classes[cls]}: {conf:.2f}"
                     cv2.putText(original_image, label, (x1, y1 - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, self.text_color, 2)
 
-        return original_image
+        print("="*80 + "\n")
+        return original_image, preprocessed_display