With　the　rapid　development　of　the　marine　economy,　the　conflict　between　the　limited　capacity　of　existing　network　infrastructure　and　the　growing　demand　for　marine　data　collection　has　become　increasingly　prominent.　Due　to　their　high　mobility　and　ease　of　deployment,　unmanned　aerial　vehicles　(UAVs)　are　playing　an　increasingly　important　role　in　marine　data　collection　and　transmission.　This　paper　investigates　the　marine　data　collection　problem　with　dual-UAV-assisted　relay.　Specifically,　it　aims　to　minimize　the　total　system　energy　consumption　while　satisfying　constraints　such　as　quality　of　service　and　service　coverage,　through　joint　optimization　of　access　relationship,　UAV　trajectory,　data　offloading,　and　communication　resource　allocation.　To　address　this　non-convex　mixed-integer　nonlinear　programming　problem,　a　joint　optimization　algorithm　is　proposed.　The　algorithm　first　employs　a　greedy　algorithm-based　approach　to　solve　the　access　relationship　optimization　sub-problem,　then　applies　a　successive　convex　approximation　(SCA)-based　approach　to　solve　the　UAV　trajectory　and　communication　resource　allocation　sub-problems,　and　finally　iteratively　solves　these　sub-problems　to　obtain　a　near-optimal　solution　to　the　original　problem.　Simulation　results　demonstrate　that,　compared　with　several　benchmark　algorithms,　the　proposed　joint　optimization　algorithm　significantly　reduces　the　total　system　energy　consumption　while　ensuring　the　successful　completion　of　all　data　collection.