Gannet　optimization　algorithm　(GOA)　is　a　meta-heuristic　algorithm　based　on　habits　of　gannet　proposed　by　Zhang　et　al.　In　this　paper,　we　propose　a　Gannet　optimization　algorithm　using　parallel　strategy　(PGOA).　Since　the　GOA　algorithm　has　the　risk　of　easily　falling　into　local　optimality,　the　use　of　the　parallel　strategy　can　largely　avoid　falling　into　local　optimality.　Therefore,　we　use　the　parallel　strategy　to　improve　the　GOA　algorithm,　which　greatly　improves　the　performance　and　efficiency　of　the　algorithm.　The　improved　algorithm　is　applied　to　image　segmentation,　and　the　processed　images　are　evaluated　using　PSNR,　SSIM,　and　FSIM　as　evaluation　metrics.　The　experimental　results　show　that　the　improved　GOA　algorithm　can　achieve　higher　quality　image　segmentation　compared　to　other　algorithms　on　image　segmentation.