番茄植株生長過程中莖葉表型的無損分割與提取
打開文本圖片集
中圖分類號：S126 文獻(xiàn)標(biāo)志碼：A 文章編號：1008-0864（2025）07-0090-11
Nondestructive Segmentation and Extraction of Stem and Leaf Phenotypes During Tomato Plant Growth
WANG Yaxin，LYU Yangcheng，WANG Wenqi，LIU Qi， YANG Jie，REN Guihong， ZHANG Wuping，LI Fuzhong* （School of Software，Shanxi Agricultural University，Shanxi Taigu O3O8O1，China）
Abstract：Aiming at thecurentproblem that itisdiffcult toextract thephenotypic parameters of tomato plants without any loss，the stem and leaf segmentation and phenotypic extraction method of tomato plants based on 3D reconstruction was proposed.Firstly，the multi-view image sequences of tomato plants were colected to construct a 3D modelof the plant，anda combinationof multiple filtering algorithms was used to complete the pre-processing. The skeletonof thepre-processed point cloud was extracted using theLaplace-based skeleton extraction algorithm， andthe segmentationof the stemand leaves was completed basedonthe skeleton of the plant，and the segmentation of the single leaveswas completedusing the method based on theclusteringof hyperbolomers.6 phenotypic parameters including the heightoftheplant，thethickness of the stem，theangleoftheinclinationoftheleaf，the lengthof theleaf，the widthoftheleaf，and theareaof theleaf werealso extracted.Theresultsshowed thatthe average accuracy，average recalland average F1 score of stemand leaf segmentation were O.88，0.8O and 0.84， respectively，andthesegmentationindexes werebeter thantheother4segmentationalgorithms.Thecoefficientsof determination between the calculatedand measured values of plant height，stem thickness，leaf inclination，leaf length，leaf width and leaf area were0.97，O.84，0.88，0.94，0.92 and0.93，respectively，and theroot mean square errors were2.17cm，0.346cm，5.65°，3.18cm，2.99 cm and 8.79cm2.The measured valuesof the proposed method hadastrongcorelation with thecalculated values，which provided technical support for high-throughput phenotypic parameter extraction in tomato plants.
Key words：tomato；three-dimensional reconstruction；stemand leaf segmentation；phenotype extraction
番茄具有較高的營養(yǎng)價值和經(jīng)濟價值，在世界各地被廣泛種植。（剩余14833字）