Random cutting plane approach for identifying volumetric features in a CAD mesh model

Muraleedharan, Lakshmi Priya; Kannan, Shyam Sundar; Karve, Ameya; Muthuganapathy, Ramanathan

doi:10.1016/j.cag.2017.07.025

“…Comparing with existing recent approaches such as Muraleedharan et al [22], Adhikary et al [14], Attene et al [12], RANSAC [16], Le and Daun [15] and other on benchmark test cases, the proposed technique successfully recognized the features such as blends, compound holes and their interactions and found to be robust and consistent with coverage of more than 95% in addressing volumetric feature. The proposed approach is simple, more general and more reliable.…”

Section: Resultsmentioning

confidence: 77%

“…Failure cases for intersecting volumetric feature Failure case of Muraleedharan et al[22], Adhikary et al[14] and Attene et.al[12] …”

mentioning

confidence: 99%

Intelligent systems for volumetric feature recognition from CAD mesh models

Verma¹,

Patil²,

Hase

³

et al. 2020

IJIE

0

View full text Add to dashboard Cite

This paper presents an intelligent technique to recognize the volumetric features from CAD mesh models based on hybrid mesh segmentation. The hybrid approach is an intelligent blending of facet-based, vertex based, rulebased, and machine learning based techniques. Comparing with existing stateof-the-art approaches, the proposed approach does not depend on attributes like curvature, minimum feature dimension, number of clusters, number of cutting planes, the orientation of model and thickness of the slice to extract volumetric features. The intelligent threshold prediction makes hybrid mesh segmentation automatic. The proposed technique automatically extracts volumetric features like blends and intersecting holes along with their geometric parameters. The proposed approach has been extensively tested on various benchmark test cases. The proposed approach outperforms the existing techniques favorably and found to be robust and consistent with coverage of more than 95% in addressing volumetric features

show abstract

“…Further, Sunil and Pande [8] have attempted to recognize hole, dimple, flange, ridge, bend while Adhikary and Gurumoorthy [20] recognized and extracted free-form volumetric features. Muraleedharan et al [21] identified holes, slots, pockets as well as interacting features CAD mesh model, but could not recognize chamfer, fillet, and blends. From a literature review, it is evident that most of the researchers have focused on extracting limited and comparatively simple features.…”

Section: Literature Findingsmentioning

confidence: 99%

“…However, their algorithm depends on the choice of "Minimum feature dimension" and must be known in advance before feature extraction. Muraleedharan et al [21] used a random cutting plane approach for feature extraction. However, their algorithm needs a number of cutting planes for FR and is assumed to be known.…”

Section: Literature Reviewmentioning

confidence: 99%

See 1 more Smart Citation

Blend recognition from CAD mesh models using pattern matching

Hase¹,

Bhalerao²,

Verma³

et al. 2019

1st International Conference on Advances in Mechanical Engineering and Nanotechnology (Icamen 2019)

View full text Add to dashboard Cite

This paper reports a unique, platform-independent approach for blend recognition from CAD mesh model using pattern matching. About 60% of the average portion of the total facets in CAD mesh model is blended features. So, it becomes essential and necessary to extract these blend features for the successful accomplishment of seamless CAD/CAM integration. The facets of the same region have similar patterns. The focus of this paper is to recognize the blends using hybrid mesh segmentation based on pattern matching. Blend recognition has been carried out in three phases viz. preprocessing, pattern matching hybrid mesh segmentation and blend feature identification. In preprocessing, the adjacency relationship is set in facets of CAD mesh model, and then Artificial Neural Networks based threshold prediction is employed for hybrid mesh segmentation. In the second phase, pattern matching hybrid mesh segmentation is used for clustering the facets into patches based on distinct geometrical properties. After segmentation, each facet group is subjected to several conformal tests to identify the type of analytical surfaces such as a cylinder, cone, sphere, or tori. In the blend feature recognition phase, the rule-based reasoning is used for blend feature extraction. The proposed method has been implemented in VC++ and extensively tested on benchmark test cases for prismatic surfaces. The proposed algorithm extracts the features with coverage of more than 95 %. The innovation lies in "Facet Area" based pattern matching hybrid mesh segmentation and blend recognition rules. The extracted feature information can be utilized for downstream applications like tool path generation, computer-aided process planning, FEA, reverse engineering, and additive manufacturing.

show abstract

Intelligent Systems for Volumetric Feature Recognition from CAD Mesh Models

Hase¹,

Bhalerao²,

Verma³

et al. 2019

EAI International Conference on Big Data Innovation for Sustainable Cognitive Computing

View full text Add to dashboard Cite

This paper presents an intelligent technique to recognize the volumetric features from CAD mesh models based on hybrid mesh segmentation. The hybrid approach is an intelligent blending of facet-based, vertex based, rulebased, and machine learning based techniques. Comparing with existing stateof-the-art approaches, the proposed approach does not depend on attributes like curvature, minimum feature dimension, number of clusters, number of cutting planes, the orientation of model and thickness of the slice to extract volumetric features. The intelligent threshold prediction makes hybrid mesh segmentation automatic. The proposed technique automatically extracts volumetric features like blends and intersecting holes along with their geometric parameters. The proposed approach has been extensively tested on various benchmark test cases. The proposed approach outperforms the existing techniques favorably and found to be robust and consistent with coverage of more than 95% in addressing volumetric features

show abstract

Random cutting plane approach for identifying volumetric features in a CAD mesh model

Cited by 14 publications

References 28 publications

Intelligent systems for volumetric feature recognition from CAD mesh models

Intelligent systems for volumetric feature recognition from CAD mesh models

Blend recognition from CAD mesh models using pattern matching

Intelligent Systems for Volumetric Feature Recognition from CAD Mesh Models

Contact Info

Product

Resources

About