Manually plot boxes, rather than using `r.plot()`

set-detect-notify/__main__.py

@@ -13,6 +13,7 @@ from ultralytics import YOLO
 import argparse
 
 from .utils import notify
+from .utils import utils
 
 DATETIME_FORMAT = "%Y-%m-%d %H:%M:%S"
 args = None
@@ -48,13 +49,13 @@ def main():
         type=float,
         help="The scale to run the detection at, default is 0.25",
     )
-    # argparser.add_argument(
+    argparser.add_argument(
-    # '--view-scale',
+    '--view-scale',
-    # # Set it to the env VIEW_SCALE if it isn't blank, otherwise set it to 0.75
+    # Set it to the env VIEW_SCALE if it isn't blank, otherwise set it to 0.75
-    # default=os.environ['VIEW_SCALE'] if 'VIEW_SCALE' in os.environ and os.environ['VIEW_SCALE'] != '' else 0.75,  # noqa: E501
+    default=os.environ['VIEW_SCALE'] if 'VIEW_SCALE' in os.environ and os.environ['VIEW_SCALE'] != '' else 0.75,  # noqa: E501
-    # type=float,
+    type=float,
-    # help="The scale to view the detection at, default is 0.75",
+    help="The scale to view the detection at, default is 0.75",
-    # )
+    )
 
     argparser.add_argument(
         "--confidence-threshold",
@@ -160,10 +161,12 @@ def main():
         # Only process every other frame of video to save time
         # Resize frame of video to a smaller size for faster recognition processing
         run_frame = cv2.resize(frame, (0, 0), fx=args.run_scale, fy=args.run_scale)
-        # view_frame = cv2.resize(frame, (0, 0), fx=`args.`view_scale, fy=args.view_scale)
+        # view_frame = cv2.resize(frame, (0, 0), fx=args.view_scale, fy=args.view_scale)
 
         results = model(run_frame, verbose=False)
-        for r in results:
+        for i, r in enumerate(results):
+            # list of dicts with each dict containing a label, x1, y1, x2, y2
+            plot_boxes = []
             # Setup dictionary of object names
             if not object_names:
                 for name in r.names.values():
@@ -174,6 +177,7 @@ def main():
                         "last_notification_time": None,
                     }
             for box in r.boxes:
+        
                 # Get the name of the object
                 class_id = r.names[box.cls[0].item()]
                 # Get the coordinates of the object
@@ -192,6 +196,18 @@ def main():
                     # If the confidence is less than 0.5, then SKIP!!!!
                     continue
 
+
+                # Add the object to the list of objects to plot
+                plot_boxes.append(
+                    {
+                        "label": class_id,
+                        "x1": cords[0],
+                        "y1": cords[1],
+                        "x2": cords[2],
+                        "y2": cords[3],
+                    }
+                )
+
                 # End goal: Send a notification when an object has been detected for 2 seconds in the past 15 seconds.
                 # However, don't send a notification if the last notification was less than 15 seconds ago
 
@@ -265,13 +281,17 @@ def main():
                         # Reset the detection duration
                         print("Just sent a notification - resetting detection duration")
                         object_names[class_id]["detection_duration"] = 0
-            im_array = r.plot()
+            frame_to_show = utils.plot_label(
-            # Scale back up the coordinates of the locations of detected objects.
+                boxes=plot_boxes,
-            # im_array = np.multiply(im_array, 1/args.run_scale)
+                full_frame=frame,
-            # print(type(im_array))
+                # full_frame=r.plot(),
-            # print(im_array)
+                run_scale=args.run_scale,
-            # exit()
+                view_scale=args.view_scale,
-            cv2.imshow("View", im_array)
+            )
+            
+            # Display the resulting frame
+            # cv2.imshow("", r)
+            cv2.imshow(f"Video{i}", frame_to_show)
 
         # Hit 'q' on the keyboard to quit!
         if cv2.waitKey(1) & 0xFF == ord("q"):

set-detect-notify/utils/utils.py

@@ -0,0 +1,46 @@
+import cv2
+import numpy as np
+def plot_label(
+        # list of dicts with each dict containing a label, x1, y1, x2, y2
+        boxes: list = None,
+        # opencv image
+        full_frame: np.ndarray = None,
+        # run_scale is the scale of the image that was used to run the model
+        # So the coordinates will be scaled up to the view frame size
+        run_scale: float = None, 
+        # view_scale is the scale of the image, in relation to the full frame
+        # So the coordinates will be scaled appropriately when coming from run_frame
+        view_scale: float = None,
+        font: int = cv2.FONT_HERSHEY_SIMPLEX,
+):
+    view_frame = cv2.resize(full_frame, (0, 0), fx=view_scale, fy=view_scale)
+    for thing in boxes:
+        cv2.rectangle(
+            # Image
+            view_frame,
+            # Start point
+            (int(thing["x1"] * (run_scale/view_scale)), int(thing["y1"] * (run_scale/view_scale))),
+            # End point
+            (int(thing["x2"] * (run_scale/view_scale)), int(thing["y2"] * (run_scale/view_scale))),
+            # Color
+            (0, 255, 0),
+            # Thickness
+            2,
+        )
+        cv2.putText(
+            # Image
+            view_frame,
+            # Text
+            thing["label"],
+            # Origin
+            (int(thing["x1"] * (run_scale/view_scale)), int(thing["y1"] * (run_scale/view_scale))),
+            # Font
+            font,
+            # Font Scale
+            1,
+            # Color
+            (0, 255, 0),
+            # Thickness
+            1
+        )
+    return view_frame