I PointNet++ : Improved pointnet++ for segmentation and localization of leather grasp points / Jin Guang in JOURNAL OF THE AMERICAN LEATHER CHEMISTS ASSOCIATION (JALCA), Vol. CXIX, N° 4 (04/2024)  

[article]  inJOURNAL OF THE AMERICAN LEATHER CHEMISTS ASSOCIATION (JALCA) > Vol. CXIX, N° 4 (04/2024) . - p. 174-183 Titre : I PointNet++ : Improved pointnet++ for segmentation and localization of leather grasp points Type de document : texte imprimé Auteurs : Jin Guang, Auteur ; Ren Gonchang, Auteur ; Huan Yuan, Auteur ; Sun Jiangong, Auteur Année de publication : 2024 Article en page(s) : p. 174-183 Note générale : Bibliogr. Langues : Américain (ame) Catégories : Cuirs et peaux Manutention -- Appareils et matériel Point de préhension Robotique Robots industriels Index. décimale : 675 Technologie du cuir et de la fourrure Résumé : In order to achieve accurate identification and positioning of leather grasp points during the process of robot grasp and spreading leather, this paper proposes a leather grasp point segmentation and positioning method based on improved PointNet++(IPointNet++). Taking leather in its natural falling state as the research object, a depth camera is used to collect point cloud data of the leather. Firstly, the preprocessing of leather point clouds is completed by removing background point clouds based on PassThroughFilter and eliminating noise based on Statistics Filter. Secondly, the octree sampling method is used to replace the farthest point sampling method of the original PointNet++, which is adapted to the nonrigid deformation characteristics of the leather itself. Thereby, the entire leather is divided into two parts: the main body and the grasp area. Lastly, the three-dimensional coordinates of the leather grasp points are obtained by solving the centroid of the point cloud data in the leather grasp area grasp. In the segmentation experiments, the improved PointNet++ has raised the mIoU by 11.8% and 2.5% comparing with PointNet and PointNet++ respectively, and the OA by 6.1% and 1.1%. In the grasp experiments, the success rate of leather grasp points identification grasp is 93.33%, and the success grasp rate grasp is 82.14%. The experimental results show that the proposed method has higher segmentation accuracy and good applicability. Note de contenu : - MATERIALS AND METHODS : Data acquisition - Point cloud preprocessing - Data Labeling - IMPROVEMENT OF THE POINTNET++ MODEL (I PointNet++) : Localization of leather frasp points - EXPERIMENTAL PROCESS AND RESULT ANALYSIS : Experimental conditions - Segmentation experiment - Grasp localization experiment - Table 1 : Experimental configuration - Table 2 : Training parameter settings - Table 3 : Segmentation accuracy results (%) - Table 4 : Test data for grasp localization experiment DOI : https://doi.org/10.34314/9y3ykt93 En ligne : https://drive.google.com/file/d/1dKFXpXeaLCElVID0fjTHz0xOudBN1eLf/view?usp=drive [...] Format de la ressource électronique : Pdf Permalink : https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=40825 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
24674	-	Périodique	Bibliothèque principale	Documentaires	Disponible

Research on experimental exploration and identification positioning algorithm for the optimal stretching point of leather in the toggling process / Huan Yuan in JOURNAL OF THE SOCIETY OF LEATHER TECHNOLOGISTS & CHEMISTS (JSLTC), Vol. 107, N° 5 (09-10/2023)  

[article]  inJOURNAL OF THE SOCIETY OF LEATHER TECHNOLOGISTS & CHEMISTS (JSLTC) > Vol. 107, N° 5 (09-10/2023) . - p. 175-182 Titre : Research on experimental exploration and identification positioning algorithm for the optimal stretching point of leather in the toggling process Type de document : texte imprimé Auteurs : Huan Yuan, Auteur ; Ren Gonchang, Auteur ; Su Xiangyu, Auteur ; Jin Guang, Auteur ; Yang Yiming, Auteur ; Zhang Heng, Auteur Année de publication : 2023 Article en page(s) : p. 175-182 Note générale : Bibliogr. Langues : Anglais (eng) Catégories : Algorithmes Cuirs et peaux -- Propriétés mécaniques Cuirs et peaux de moutons Etirement (mécanique) Index. décimale : 675 Technologie du cuir et de la fourrure Résumé : The objective of this research is to address the issue of automated selection of the optimal stretching position for leather during the leather toggling process. By analysing a substantial number of leather profiles and the actions involved in manual stretching and leather stretching, it has been determined that selecting appropriate stretching positions and intervals is the pivotal factor in enhancing leather yield. Through orthogonal experimentation, it has been deduced that the ideal distances between the stretching points on both sides of the head and tail, as well as on both sides of the body parallel to the midrib line, are 12cm and 9cm respectively. Building upon this, the study proposes an algorithm for identifying and locating the most suitable stretching positions based on the simplified skeleton method and the intersection method of drawn circles. The algorithm employs machine vision to obtain the initial image of the leather source, followed by preprocessing to eliminate image defects and accentuated contours. Subsequently, the thinning algorithm extracts the image skeleton, and the convolution operation filters out the ridges and limbs of the leather, resulting in the simplest skeleton that determines the requisite stretching points. Building upon the determination of the necessary stretching points, the remaining contour regions of the leather are established using the circle intersection method, thus obtaining the remaining stretching points. The identified algorithm for identifying the optimal stretching points in this study holds significant importance in the research and development of automated equipment for the leather stretching plate drying process, and serves as a valuable reference for enhancing leather yield and optimising the process. Note de contenu : - The method and principle of selecting stretching points - Exploration experiment of optimal stretching points - Design of the optimal stretch point identification algorithm : Image preprocessing - Selection of indispensable stretching points - Selection of other stretching ponts - ALgorithm verification and result analysis - Table 1 : Table of horizontal parameters of various factors - Table 2 : Orthogonal experiment scheme and result record table - Table 3 : Range analysis of the experimental results En ligne : https://drive.google.com/file/d/1rir9LlaHepJ3AlnIpukUQ58D3AOk_wiA/view?usp=share [...] Format de la ressource électronique : Pdf Permalink : https://e-campus.itech.fr/pmb/opac_css/index.php?lvl=notice_display&id=39956 [article]

Réservation

Réserver ce document

Exemplaires (1)

Code-barres	Cote	Support	Localisation	Section	Disponibilité
24238	-	Périodique	Bibliothèque principale	Documentaires	Disponible

---

1 (1 - 2 / 2)