基于改進(jìn)DBSCAN和距離共識(shí)評(píng)估的分段點(diǎn)云去噪方法
系統(tǒng)仿真學(xué)報(bào)
頁數(shù): 10 2024-07-03
摘要: 針對(duì)點(diǎn)云數(shù)據(jù)中噪聲點(diǎn)的剔除問題,提出了一種基于改進(jìn)DBSCAN(density-based spatial clustering of applications with noise)算法的多尺度點(diǎn)云去噪方法。應(yīng)用統(tǒng)計(jì)濾波對(duì)孤立離群點(diǎn)進(jìn)行預(yù)篩選,去除點(diǎn)云中的大尺度噪聲;對(duì)DBSCAN算法進(jìn)行優(yōu)化,減少算法時(shí)間復(fù)雜度和實(shí)現(xiàn)參數(shù)的自適應(yīng)調(diào)整,以此將點(diǎn)云分為正常簇、疑似簇及異常簇,并...