If your journal is easily accessible then I would NOT re-use the image, but instead refer to the exact place in the earlier paper. If everyone re-hashed their graphs or tables then we would be flooded eventually - it's important indeed to describe earlier work that you build your new paper on, but an image? No, refer to the previous paper. I would also assume that the publisher would not want to print an already existing image, as it means that there is less space in that journal issue for new things.

Thank you professor Michael Patriksson

You're welcome! :-)