In　pipeline　transportation,　which　offers　advantages　such　as　safety　and　efficiency　when　transporting　gases,　liquids,　and　other　substances,　regular　inspection　and　maintenance　of　pipelines　are　of　paramount　importance　to　ensure　their　safe　and　uninterrupted　operation.　In　the　course　of　transportation,　pipelines　may　experience　deformation　due　to　both　intrinsic　factors　and　external　influences,　or　they　may　become　obscured　or　buried　as　a　result　of　non-human-induced　external　factors.　Such　occurrences　can　lead　to　the　leakage　of　substances　within　the　pipelines,　resulting　in　significant　economic　losses　and　environmental　contamination.　Ensuring　the　safety　and　unobstructed　operation　of　oil　transportation　pipelines　is　of　paramount　importance.This　paper　presents　a　pipeline　target　identification　and　displacement　monitoring　approach　based　on　unmanned　aerial　vehicle　(UAV)　flight　trajectories　and　known　information.　It　involves　receiving　and　streaming　data　collected　from　the　onboard　UAV　system,　cropping　real-time　1080p　image　data,　inputting　it　into　the　ENet　network　for　road　semantic　segmentation,　and　preprocessing　the　ENet　network＇s　output　target　region　images　with　connected　component　analysis　and　histogram　enhancement.　These　preprocessed　images　are　then　input　into　a　traditional　network　algorithm　model　for　straight-line　detection.　Finally,　using　the　pipeline　displacement　detection　method　based　on　the　onboard　platform＇s　recognition　and　positioning　algorithms,　the　detected　results　for　pipeline　displacement　information　are　obtained.Through　testing　and　validation,　this　proposed　method　in　the　paper　demonstrates　reduced　data　requirements,　higher　real-time　capability,　and　better　alignment　with　the　ground-based　integrated　processing　platform　system＇s　needs　for　pipeline　displacement　detection.　©　2023　ACM.