A Study On Volume Segmentation Algorithm For CT Images Of Plant Organs

As the most common scenery on the earth,plants are integral part of the earth's ecosystems.Over the years,researchers have been exploring the mysteries of plant life and its growth process and they have achieved some results.Flowers are one of the most important organs of the plant.They are not only diverse in variety,but also possess complex morphological structures and life characteristics.The building of the high-precision and high-realistic flower object model is still one of the research hotspots.However,due to the complex structures of the flowers and the tightness of the internal tissues,the data field segmentation of three-dimensional plant organs has always been a challenging subject.To solve the problem of segmentation and visualization of CT images of plant organs.This study mainly focused on the flowers.The X-ray computed tomography system(CT)was used to obtain the information of the flowers' body data,and the obtained flowers' body data information was segmented and the skeletons of the plant organ image are extracted by morphological refinement algorithm.And then a hit-miss transform is adopted to obtain the endpoint information of the feature,thereby providing data support for the surface reconstruction of the plant organ.The experimental results show that this algorithm can more accurately extract the volume data of flowers and achieve better results in the subsequent visual processing.It has important application value.This article focuses on the study of the segmentation of the flower organ CT images and the subsequent extraction of skeletons and endpoint information of feature.The main tasks are as follows:(1)Study on the segmentation algorithm of the flower organ CT images.Although the CT system is capable of capturing body data of the flowers' internal structure,it is still difficult to segment them into specific organs accurately and model them as smooth curvedsurfaces.Because the flowers are composed of thin petals that touch each other.Combining the MeanShift algorithm and the Graph Cut algorithm,this paper reconstructs the energy function data item and the smooth item of the classical graph cutting algorithm,and then uses the max-flow/min-cut algorithm to process the segmentation target.(2)The feature extraction and visualization of plant organ.This part of the work was carried out on the basis of previous work.After the CT segmentation of the flower organ was completed,the skeleton was extracted from each segmented plant organ image by discrete curve evolution and morphology refinement algorithms.The hit-miss transform is adopted to extract feature endpoints and the extracted skeleton and feature endpoint information is saved as feature point data.Next,we make visualization processing on the point cloud data through the linear octree tree model and display the obtained point cloud data visually and realize viewing the internal point cloud information and the extracted feature point data information interactively.