I have a gazebo simulation setup with multiple stereo camera. The stereo camera has their topic set to publish in the following format:

stereo_1/left/image_raw
stereo_1/right/image_raw
stereo_2/left/image_raw
stereo_2/right/image_raw

I am using a package that calls upon the opencv_app edge detection node:

<node name="edge_detection" pkg="opencv_apps" type="edge_detection" output="$(arg stdout)" >
    <remap from="image" to="$(arg camera_name)/left/$(arg image_topic)" />
    <rosparam>
      use_camera_info: False
      debug_view: False
      edge_type: 0
      canny_threshold1: 100
      canny_threshold2: 200
      apertureSize: 3
      apply_blur_pre: True
      postBlurSize: 13
      postBlurSigma: 3.2
      apply_blur_post: False
      L2gradient: False
    </rosparam>
  </node>

How can I set which stereo camera does the edge detection get its image from? Currently it gets it only from the stereo1/left/imageraw.