Near Optimal Approach to General Manufacturing Feature Extraction

1995 ◽  
Author(s):  
Jian (John) Dong ◽  
Sreedharan Vijayan
Keyword(s):  
Feature Extraction ◽  
Process Planning ◽  
Planning System ◽  
Manufacturing Cost ◽  
Design Data ◽  
Feature Extraction Method ◽  
Automatic Feature Extraction ◽  
Starting Point ◽  
Planning Function ◽  
Optimal Approach

Abstract Computers are being used increasingly in the process planning function. The starting point of this function involves interpreting design data from a CAD model of the designed component Feature-based technology is becoming an important tool for this. Automatic recognition of features and extraction of feature information from CAD data can be used to drive a process planning system. In this paper a new approach to automatic feature extraction called the Blank-Surface Concave-edge (BS-CE) approach is illustrated. This approach attempts to remove as much of the blank material with a given machine setup as possible. Hence intuitively one can say that the manufacturing cost of material removal may be minimized if this technique is employed. This feature extraction method is explained along with examples of its implementation. An analysis of alternate feature extraction results is performed and the cost of manufacture is compared to demonstrate the near optimal performance of this technique.

Feature Extraction Techniques by Optimal Volume Decomposition

1996 ◽  
Author(s):  
Jian Dong ◽  
Sreedharan Vijayan
Keyword(s):  
Feature Extraction ◽  
Process Planning ◽  
Feature Recognition ◽  
Optimization Methods ◽  
Manufacturing Cost ◽  
Extraction Techniques ◽  
Manufacturing Planning ◽  
Cad Cam ◽  
Feature Based ◽  
Manufacturing Feature

Abstract The elements of Computer-Aided Manufacturing, do not make full use of the part description stored in a CAD model because it exists in terms of low-level faces, edges and vertices or primitive volumes related to the manufacturing planning task. Consequently manufacturing planning still depends upon human expertise and input to interpret the part definition according to manufacturing needs. Feature-based technology is becoming an important tool to resolve this and other related problems. One approach is to design the part using Features directly. Another approach is Manufacturing Feature Extraction and Recognition. Manufacturing Feature Extraction consists of searching for the part description, recognizing cavity features, extracting those features as solid volumes of material to be removed. Feature Recognition involves raising this information to the level of part features which can be read by a process planning program. The feature extraction can be called optimal if the manufacturing cost of the component using those features can be minimized. An optimized feature extraction technique using two powerful optimization methods viz., Simulated Annealing and Genetic Algorithm is presented in this paper. This work has relevance in the areas of CAD/CAM linking, process planning and manufacturability assessment.

An Attempt to Application of Chain Codes for Design Similarity Evaluation

2014 ◽  
Vol 1036 ◽  
pp. 897-902 ◽  
Author(s):  
Cezary Grabowik ◽  
Krzysztof Kalinowski ◽  
Iwona Paprocka ◽  
Wojciech Kempa
Keyword(s):  
Process Planning ◽  
Edit Distance ◽  
Hamming Distance ◽  
Planning System ◽  
Similar Process ◽  
Computer Aided Process Planning ◽  
Chain Code ◽  
Process Plans ◽  
Chain Codes ◽  
Starting Point

The need of integration of all engineering activities starting from a designing stage finishing on manufacturing implies the necessity of application of a new software solution. Among them CAPP systems seem to be the best and the most promising integration tool. There are the two basic methods of the computer aided process planning it is the variant and generative methods. These CAPP modelling methods are based on the different principles. The variant method is based, in general, on the ideas of the design similarity and group technology whilst the generative one mainly on the basis of utilization of the manufacturing knowledge store and the automatic process synthesis. It is a well-known fact that the variant process planning approach is based on the hypothesis that similar parts would have similar process plans. In a CAPP variant process planning system it is necessary to look for similarities between parts. Process plans formerly prepared and stored in the CAPP system manufacturing database are used for planning of new manufacturing processes for new parts similar to those recorded in system database. In the variant system working out of the classification schema and similarity measures is essential. In this paper an attempt to application of chain codes for a design similarity evaluation is presented. The proposed procedure is based on application of the chain codes for description of the parts selected longitudinal cross sections. The proposed chain code structure is created in similar way to this used for the Freeman code formation but unlike classic Freeman chain code the parts cross section is discretized in more than the 8 cardinal directions. The characteristic feature of the proposed profile coding method is the approach for the code starting point selection. In each case the part design chain code is created starting from the a priori selected upper left corner of a product. The idea of products similarity evaluation bases on the parts profile codes and application of the minimum edit distance definitions. In the paper the Levenshtein and Hamming distance for code chains similarity comparison are proposed At the current stage of research this method is used for similarity evaluation of rotational symmetric parts. The choice of the minimum edit distance measure in the proposed solution depends on product pins number.

scholarly journals Ranked Features Selection with MSBRG Algorithm and Rules Classifiers for Cervical Cancer

2019 ◽  
Vol 15 (12) ◽  
pp. 4
Author(s):  
Mohammad Subhi Al-batah
Keyword(s):  
Cervical Cancer ◽  
Feature Extraction ◽  
Feature Selection ◽  
Automatic Segmentation ◽  
Region Growing ◽  
Training Data ◽  
Decision Table ◽  
Features Selection ◽  
Feature Extraction Method ◽  
Automatic Feature Extraction

<p class="0abstract">In this paper, an automatic three-phase cervical cancer diagnosis system is employed which includes feature extraction, feature selection followed by classification. Firstly, the modified seed-based region growing (MSBRG) algorithm is implemented for automatic segmentation and feature extraction using 500 cervical cancer cells. Processes to obtain the threshold values and the initial seed location are carried out automatically using moving k-mean (MKM) algorithm and invariant moment techniques. Secondly, eight attribute evaluators are applied for selecting and ranking the features, which are Correlation-based Feature Selection, Classifier Attribute Evaluator, Correlation Attribute Evaluator, Gain Ratio, Info Gain, OneR, ReliefF, and Symmetrical Uncertainty. Finally, the classification is compared based on five classifiers: Decision Table, JRip, OneR, PART, and ZeroR. The performance of the classifiers is evaluated using 3 test options: the training percentage splits (50% to 98%), the full training data and the cross validation (2-fold to 10-fold). The experimental results prove the capability of the MSBRG algorithm as an automatic feature extraction method. Furthermore, this paper proves the ability of the ranked feature selection methods to select important features of a cervical cell, and favors the Decision Table as the best classifier for cervical cancer classification.</p>

scholarly journals An automatic feature extraction method and its application in fault diagnosis

2017 ◽  
Vol 19 (4) ◽  
pp. 2521-2533 ◽  
Author(s):  
Shunming Li ◽  
Jinrui Wang ◽  
Xingxing Jiang ◽  
Chun Cheng
Keyword(s):  
Feature Extraction ◽  
Fault Diagnosis ◽  
Extraction Method ◽  
Feature Extraction Method ◽  
Automatic Feature Extraction

A Rapid and Automatic Feature Extraction Method for Artificial Targets Used in Industrial Photogrammetry Applications

2012 ◽  
Vol 170-173 ◽  
pp. 2995-2998
Author(s):  
Jian Wei Liu ◽  
Zhi Qiang Jiang ◽  
Hao Hu ◽  
Xin Yin
Keyword(s):  
Feature Extraction ◽  
Reference Point ◽  
Extraction Method ◽  
Target Location ◽  
Reference Points ◽  
Improved Method ◽  
Feature Extraction Method ◽  
Automatic Feature Extraction ◽  
Centroid Algorithm ◽  
Point Detection

Distributing artificial targets on the object to be measured is a reliable and common method for achieving optimum target location and accurate correspondence among multi-view images, which are universally adopted in industrial photogrammetry applications. In this paper artificial circular un-coded targets and coded targets are used as reference points, an automatic and rapid algorithm for reference point detection is proposed. Targets are extracted from the images according to their size, shape, intensity , etc. An improved method to identify the ID of the coded target is proposed. The gray scale centroid algorithm is applied to get sub-pixel locations of both un-coded and coded targets. Practical examples show that the algorithm can identify and locate artificial targets in images quickly and accurately. It is robust to the change of projection angles and noise.

Sign in / Sign up

Export Citation Format

Share Document