Title: 應用圖形識別於大量客製化生產流程之研究-以工業電腦業者G公司為例
A Study of Applying Pattern Recognition into Mass Customization Manufacturing Process- A Case of an Industrial PC Company
Authors: 游世豪
林君信
管理學院管理科學學程
Keywords: 圖形辨識;客製化生產;工業電腦;pattern recognition;customization production;industrial PC
Issue Date: 2010
Abstract:   由於消費者意識的抬頭,以及數位網路帶來的各種商機,導致生產模式的改變,使得「大量客製化」這個新名詞最近廣受全球產業矚目,它是以顧客為導向的生產模式;要成功的達成大量客製化,需要選擇適當的供應鏈策略,同時搭配其它相關的輔助工具,如模組化設計、產線自動化管理等。 本論文提出以形態學為基礎的方法來定位條碼標籤,條碼標籤辨識系統由三部分構成,首先,利用形態學基礎的方法來抽取具高對比特徵的標籤;接著,如果標籤被切割幾個部分,再利用校正演算法予以校正;最後將字元切割並使用樣板比對與結構分析法來辨識字元。   本論文由工業型電腦生產製造管理開始了解,由於其少量多樣化的特性,在生產流程的控管極為重要。本研究藉由圖形識別技術來建構一個生產流程系統,利用原本的條碼標籤,提出圖形識別系統的設置與運作概念架構,及其導入模式,並加上中介層和應用層的資訊系統架構,使得每項生產中的元件或半成品均可以成為資訊系統可以追蹤控管的物件,以增加自動化生產的效率。
Since the rise of consumer awareness, and digital network society for a variety of opportunities, lead to the production patterns change, making the new term "mass customization" got global industry attention. It is a customer-oriented production model. In order to successfully achieve mass customization, it needs to select the appropriate supply chain strategy, and with other relevant instruments, such as modular design, production line automation and management. This thesis presents a morphology-based method for extracting barcode labels from cluttered images. The recognition system consists of three major components. At the first, a morphology-based method is proposed to extract important contrast features as guides to search the desired barcode labels. A recovery algorithm is then applied for reconstructing a barcode label if the label is fragmented into several parts. The last step of the proposed method performs character division and use template matching and structural analysis method to recognize the alphanumeric codes. This thesis by the industry-based computer manufacturing management began to understand, because of its small amount of diversity, production process control is extremely important. This research apply pattern recognition technology to construct a production workflow system, using the original bar code label, propose pattern recognition system setup and operation concept of architecture, its implementation model, and with middleware and application layer information system architecture. In order to increase the efficiency of automated production, it makes each component or semi-finished products in the production can be controlled and tracked by information systems.
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT079762506
http://hdl.handle.net/11536/46180
Appears in Collections:Thesis


Files in This Item:

  1. 250601.pdf

If it is a zip file, please download the file and unzip it, then open index.html in a browser to view the full text content.