In The segmenTaTion of elecTronic componenTs under The laser poinT cloud，The complex siTuaTion of disTribuTion be- Tween The workpieces makes iTimpossible for The TradiTional poinTcloud segmenTaTion algoriThm To achieve correcTsegmenTaTion and counTing. In This paper，a poinT cloud segmenTaTion meThod using LCCP and euclidean clusTering is proposed. The LCCP algoriThm based on surface convexiTy is used To perform coarse segmenTaTion of poinT clouds，separaTe workpieces sTacked on Top of each oTher wiTh small collision area , and exTracTThe poinTcloud clusTers of The remaining colliding workpieces；Secondly，The boundary deTecTion meThod based on normal esTimaTion is used To remove The edge poinTpoinTs of The exTracTed poinTcloud clusTers，so ThaTThe poinT clouds of The workpieces in conTacT wiTh each oTher are separaTed in space，and Then The euclidean clusTering algoriThm is used for fine segmenTaTion；Finally，The resulTof The coarse and fine segmenTaTion is merged To compleTe The segmenTaTion. The resulTs show ThaTunder complex siTuaTions，The proposed meThod can accuraTely spliTThe TargeTarTifacT, and The accuracy of The segmenTaTion is improved by 35. 9%，66. 7% and 80% compared wiTh LCCP, regional growTh and Eu- clidean clusTering algoriThms，respecTively，and The average Time consumpTion per 10 TargeTs is abouT290 ms，which has good accuracy and pracTicaliTy.