yolo_encoding param added

Author: mgonzs13

README.md

@@ -125,6 +125,7 @@ These are the parameters from the [yolo.launch.py](./yolo_bringup/launch/yolo.la
 - **model**: YOLO model (default: yolov8m.pt)
 - **tracker**: tracker file (default: bytetrack.yaml)
 - **device**: GPU/CUDA (default: cuda:0)
+- **yolo_encoding**: Encoding to convert input image before using YOLO (default: bgr8)
 - **enable**: whether to start YOLO enabled (default: True)
 - **threshold**: detection threshold (default: 0.5)
 - **iou**: intersection Over Union (IoU) threshold for Non-Maximum Suppression (NMS) (default: 0.7)

yolo_bringup/launch/yolo.launch.py

@@ -57,6 +57,13 @@ def run_yolo(context: LaunchContext, use_tracking, use_3d):
             description="Device to use (GPU/CPU)",
         )
 
+        yolo_encoding = LaunchConfiguration("yolo_encoding")
+        yolo_encoding_cmd = DeclareLaunchArgument(
+            "yolo_encoding",
+            default_value="bgr8",
+            description="Encoding of the input image topic",
+        )
+
         enable = LaunchConfiguration("enable")
         enable_cmd = DeclareLaunchArgument(
             "enable",
@@ -229,6 +236,7 @@ def run_yolo(context: LaunchContext, use_tracking, use_3d):
                     "model_type": model_type,
                     "model": model,
                     "device": device,
+                    "yolo_encoding": yolo_encoding,
                     "enable": enable,
                     "threshold": threshold,
                     "iou": iou,
@@ -295,6 +303,7 @@ def run_yolo(context: LaunchContext, use_tracking, use_3d):
             model_cmd,
             tracker_cmd,
             device_cmd,
+            yolo_encoding_cmd,
             enable_cmd,
             threshold_cmd,
             iou_cmd,

yolo_ros/yolo_ros/yolo_node.py

@@ -55,6 +55,9 @@ def __init__(self) -> None:
         self.declare_parameter("model_type", "YOLO")
         self.declare_parameter("model", "yolov8m.pt")
         self.declare_parameter("device", "cuda:0")
+        self.declare_parameter("yolo_encoding", "bgr8")
+        self.declare_parameter("enable", True)
+        self.declare_parameter("image_reliability", QoSReliabilityPolicy.BEST_EFFORT)
 
         self.declare_parameter("threshold", 0.5)
         self.declare_parameter("iou", 0.5)
@@ -66,9 +69,6 @@ def __init__(self) -> None:
         self.declare_parameter("agnostic_nms", False)
         self.declare_parameter("retina_masks", False)
 
-        self.declare_parameter("enable", True)
-        self.declare_parameter("image_reliability", QoSReliabilityPolicy.BEST_EFFORT)
-
         self.type_to_model = {"YOLO": YOLO, "World": YOLOWorld}
 
     def on_configure(self, state: LifecycleState) -> TransitionCallbackReturn:
@@ -80,6 +80,9 @@ def on_configure(self, state: LifecycleState) -> TransitionCallbackReturn:
         )
         self.model = self.get_parameter("model").get_parameter_value().string_value
         self.device = self.get_parameter("device").get_parameter_value().string_value
+        self.yolo_encoding = (
+            self.get_parameter("yolo_encoding").get_parameter_value().string_value
+        )
 
         # inference params
         self.threshold = (
@@ -326,8 +329,9 @@ def image_cb(self, msg: Image) -> None:
         if self.enable:
 
             # convert image + predict
-            cv_image = self.cv_bridge.imgmsg_to_cv2(msg, desired_encoding="bgr8")
-            cv_image = cv2.cvtColor(cv_image, cv2.COLOR_BGR2RGB)
+            cv_image = self.cv_bridge.imgmsg_to_cv2(
+                msg, desired_encoding=self.yolo_encoding
+            )
             results = self.yolo.predict(
                 source=cv_image,
                 verbose=False,