Two manufacturing applications of the fuzzy K-NN algorithm

✍ Scribed by Liao T. Warren; Li Damin

• Book ID: 104292131
• Publisher: Elsevier Science
• Year: 1997
• Tongue: English
• Weight: 998 KB
• Volume: 92
• Category: Article
• ISSN: 0165-0114
• DOI: 10.1016/s0165-0114(96)00176-5

✦ Synopsis

This paper discusses the applications of the fuzzy K-NN (K-nearest neighbors) algorithm for identifying welds from digitized radiographic images and for determining PCBN (polycrystalline cubic boron nitride) tool failure in face milling operations. Both applications consist of two major steps: feature extraction and pattern classification. For the weld identification application, the weld image is processed line by line and three features are extracted for each object in the line image. These features are the width, the mean square error (MSE) between the object and its Gaussian, and the peak intensity (gray level). For the application of tool failure recognition, two features are derived from AE signals generated by the cutting operation. They are the ARMS and peak/count ratio. The use of the fuzzy K-NN classifier and the classification results are discussed. The results show that the fuzzy K-NN classifier yields high successful rates of recognition for both applications.