A developer is tasked with automating a process that interacts with a remote application via a Citrix environment. The automation must extract specific fields from a non-standard form layout where traditional selectors are ineffective. After implementing a solution using AI Computer Vision, testing reveals that the accuracy of field detection varies significantly depending on the screen resolution of the remote session. Which combination of actions is the most effective strategy to create a resolution-independent and robust automation?