影像自動分群應用於工具機路徑規劃之研究

NCUIR > Engineering College > Graduate Institute of Mechanical Engineering > Electronic Thesis & Dissertation > Item 987654321/84318

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/84318

Title:	影像自動分群應用於工具機路徑規劃之研究
Authors:	林柔廷;Lin, Jou-Ting
Contributors:	機械工程學系
Keywords:	影像處理;模糊理論;路徑規劃
Date:	2020-07-23
Issue Date:	2020-09-02 19:02:49 (UTC+8)
Publisher:	國立中央大學
Abstract:	近年來，機器視覺在工業界應用的地方越來越多，尤其在材料切割與雕刻上，其輸入多為圖像，為了使輸入影像符合使用者需求，常需使用者事先對圖像進行前處理，再輸入至機台進行加工，增加使用上的困難，且一般工具機加工多為 X-Y 方向，若圖像較複雜，其加工路徑可能並未是最佳路徑，造成時間成本，因此本研究希望融合影像處理與路徑規劃，提出新的路徑規劃法。本研究使用 C#作為開發語言，搭配機器學習與模糊理論，使影像可自動分群並進行切割，以符合使用者需求，並配合輪廓跟蹤產生有別於一般 X-Y 方向的路徑規劃，並探討如分群演算法之利弊、影像處理之效果和路徑規劃之優劣等問題。關鍵字：影像處理、模糊理論、路徑規劃;In recent years, there are more and more applications of machine vision in the industrial world, especially in material cutting and engraving, the input is mostly images. In order to make the input image meet the needs of users, users often need to pre-image Processing, and then input to the machine for processing, increasing the difficulty of use, and the general tool machining is mostly in the XY direction. If the image is more complicated, the processing path may not be the best path, resulting in time cost, so this study Hope to combine image processing and path planning, and propose a new path planning method. This research uses C# as the development language, combined with machine learning and fuzzy theory, so that the images can be automatically grouped and cut to meet the needs of users, and the contour tracking generates a path plan that is different from the general XY direction, and discusses such as the pros and cons of Cluster Analysis, the effect of image processing and the advantages and disadvantages of path planning. Keyword：Image processing、Fuzzy Theory、Route Plan
Appears in Collections:	[Graduate Institute of Mechanical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	83	View/Open

社群 sharing

Loading...