The influence of these images on the accuracy of the output will depend on their location:
1 ) if they are spread at different places in the project area and each of these inaccurate images is surrounded by accurate images, the software will compensate for it and it should not have an influence.
2 ) if all the inaccurate images are grouped at the same spot (e.g. corner of the project), the accuracy of this area will probably be lower and we would trust it less. In this case, the accuracy information would have been lost during the flight.
What is important, is that the software knows that these images have a lower accuracy as seen in your screenshot and that there is enough overlap between the images, so that it can correct for it.