查詢結果分析
來源資料
頁籤選單縮合
題 名 | 以Fuzzy C-Means聚類法進行梭織物組織自動化辨識之研究=Automatic Recognition of Weave Patterns by Fuzzy C-Means Clustering Method |
---|---|
作 者 | 施中揚; 張騏麟; 賴書蓉; | 書刊名 | 紡織中心期刊 |
卷 期 | 13:4 2003.10[民92.10] |
頁 次 | 頁329-335 |
分類號 | 478.11 |
關鍵詞 | FCM聚類演算法; 織物組織; FCM clustering algorithm; Fabric weave pattern; |
語 文 | 中文(Chinese) |
中文摘要 | 本研究提出一個非監督式的梭織物組織辨識法。首先以彩色掃描器來取得單色織物的灰階數位影像,然後根據織物的灰階值統計拔出經緯紗交錯位置,再由一階及二階的統計法計算經緯紗交錯區域的紋理特性,而經緯浮的判別準則是以Fuzzy C-Means (FCM) 聚類的方法進行。由實驗結果顯示梭織物的織物組織能辨識的很好。 |
英文摘要 | A new unsupervised method is proposed for fabric weave pattern recognition. The grey-level image of a solid woven fabric is captured by a color scanner. Based on grey-level value statistics, warp and weft crossed areas are located. The texture features of warp and weft crossed areas are obtained by first-order and second-order statistics method. The decision rules for recognizing warp and weft floats are developed on the fuzzy c-means clustering method. Experimental results demonstrate that weave patterns of fabrics can be identified well. |
本系統中英文摘要資訊取自各篇刊載內容。