Update app.py

app.py

@@ -75,11 +75,10 @@ def custom_nms(preds, iou_threshold=0.3):
             filtered_preds.append(pred)
     return filtered_preds
 
-# Note the added multiplier parameter.
 def process_image(job_id, image_path, object_type, multiplier):
     try:
         jobs[job_id]['progress'] = 10
-        # Load image
+        # Load original image
         image = cv2.imread(image_path)
         if image is None:
             jobs[job_id]['progress'] = 100
@@ -97,18 +96,37 @@ def process_image(job_id, image_path, object_type, multiplier):
                 jobs[job_id]['progress'] = 100
                 jobs[job_id]['result'] = {"error": "Roboflow model not available."}
                 return
-            try:
-                results = box_model.predict(image_path, confidence=50, overlap=30).json()
-                predictions = results.get("predictions", [])
-            except Exception as e:
-                jobs[job_id]['progress'] = 100
-                jobs[job_id]['result'] = {"error": "Error during Roboflow prediction."}
-                return
 
+            # Upscale the image if it's small so that small boxes become more prominent.
+            scale_factor = 1  # default: no upscaling
+            if img_width < 1000 or img_height < 1000:
+                scale_factor = 2  # adjust factor as needed
+
+            # Run detection on the appropriate image version.
+            if scale_factor > 1:
+                upscaled_image = cv2.resize(image, None, fx=scale_factor, fy=scale_factor, interpolation=cv2.INTER_LINEAR)
+                temp_path = "upscaled.jpg"
+                cv2.imwrite(temp_path, upscaled_image)
+                results = box_model.predict(temp_path, confidence=50, overlap=30).json()
+            else:
+                results = box_model.predict(image_path, confidence=50, overlap=30).json()
+            
+            predictions = results.get("predictions", [])
             processed_preds = []
             for prediction in predictions:
                 try:
-                    x, y, width, height = prediction["x"], prediction["y"], prediction["width"], prediction["height"]
+                    if scale_factor > 1:
+                        # Adjust coordinates from the upscaled image back to the original scale.
+                        x = prediction["x"] / scale_factor
+                        y = prediction["y"] / scale_factor
+                        width = prediction["width"] / scale_factor
+                        height = prediction["height"] / scale_factor
+                    else:
+                        x = prediction["x"]
+                        y = prediction["y"]
+                        width = prediction["width"]
+                        height = prediction["height"]
+
                     x1 = int(round(x - width / 2))
                     y1 = int(round(y - height / 2))
                     x2 = int(round(x + width / 2))
@@ -125,10 +143,11 @@ def process_image(job_id, image_path, object_type, multiplier):
                 except Exception as e:
                     continue
 
+            # Apply Non-Maximum Suppression.
             box_detections = custom_nms(processed_preds, iou_threshold=0.3)
             jobs[job_id]['progress'] = 60
 
-            # (Optional) Detect ArUco marker for measurement
+            # (Optional) Detect ArUco marker for measurement.
             marker_real_width_cm = 10.0  
             try:
                 gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
@@ -217,9 +236,11 @@ def process_image(job_id, image_path, object_type, multiplier):
         detection_counts = Counter(det["class"] for det in detection_info)
         if detection_counts:
             top_text = ", ".join(f"{cls}: {count}" for cls, count in detection_counts.items())
-            (info_width, info_height), info_baseline = cv2.getTextSize(top_text, cv2.FONT_HERSHEY_SIMPLEX, 1, 2)
+            summary_font_scale = 1 * multiplier
+            summary_thickness = 2 * multiplier
+            (info_width, info_height), info_baseline = cv2.getTextSize(top_text, cv2.FONT_HERSHEY_SIMPLEX, summary_font_scale, summary_thickness)
             cv2.rectangle(image, (5, 5), (5 + info_width, 5 + info_height + info_baseline), (0, 255, 0), -1)
-            cv2.putText(image, top_text, (5, 5 + info_height), cv2.FONT_HERSHEY_SIMPLEX, 1, (0, 0, 0), 2)
+            cv2.putText(image, top_text, (5, 5 + info_height), cv2.FONT_HERSHEY_SIMPLEX, summary_font_scale, (0, 0, 0), summary_thickness)
 
         jobs[job_id]['progress'] = 100
         retval, buffer = cv2.imencode('.jpg', image)
@@ -500,7 +521,6 @@ def analyze():
     object_type = request.form.get('object_type', '').lower()
     if object_type not in {"person", "car", "box"}:
         return jsonify({"error": "Invalid object type."}), 400
-    # Retrieve the multiplier sent from the client
     try:
         multiplier = int(request.form.get('multiplier', 1))
     except ValueError: