English  |  正體中文  |  简体中文  |  Items with full text/Total items : 51296/86402 (59%)
Visitors : 8164773      Online Users : 90
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: http://tkuir.lib.tku.edu.tw:8080/dspace/handle/987654321/103134


    Title: Genetic Programming於影像處理應用技術之探討
    Other Titles: A Study of Image Processing Techniques with Genetic Programming
    Authors: 張峯誠
    Contributors: 淡江大學資訊創新與科技學系
    Keywords: 演化式計算;基因程式法;影像處理;訊號處理;evolutionary computation;genetic programming;image processing;signal processing
    Date: 2012-08
    Issue Date: 2015-05-19 15:23:49 (UTC+8)
    Abstract: 由於計算機硬體的快速進步,使得演化式計算(Evolutionary Computation, EC)應 用逐漸擴張。EC 的基本精神是接受環境檢驗的個體,除了由親代的特徵以機率排列外, 也有部分特徵可能以特定機率發生突變。這種引入機率概念的變化,將使整個群體產生 多樣性,當環境變動時,隨著世代演進,個體特徵將逐漸傾向較適合生存的特徵組合。 而當環境不變時,整個群體可能逐漸向次佳特徵收斂,不過由於突變的關係,一直會有 偏離收斂點的子代產生,因此整群體一直保持有機會跳脫現狀,並產生最佳個體的狀態。 本計畫「Genetic Programming 於影像處理應用技術之探討」的目的,在探討適合 Genetic Programming(或GP 結合其他Evolutionary Computation 方法)應用於影像資料 的處理架構。在本研究中,我們將探討下列幾個方面:(1) 熟悉GP 之實際應用方法, 並設計適合影音處理時,降低GP 運算複雜度的技巧;(2) 影像處理時往往有許多可平 行化的地方,與GP 的特徵類似,將設計一個軟體架構,適合整合GP 方法於多核心甚 至是網路環境中提高運算效能;(3) 以實際應用驗證GP 的效果,並與傳統作法進行比 較。 本研究試圖結合GP 與影像處理,研究內容不論在學理與應用方面,過程中對相關 技巧的探討與設計、研究模擬產出的演算法等,這些成果將可作為其他相關研究的重要 參考。
    With the fast advances in computer hardware, evolutionary computation (EC) becomes feasible for various kinds of applications. The fundamental concept of EC is to introduce probability when searching the desired solution. The concept could be implemented as the operation of generating evolved solutions. For example, the crossover operation randomly switches the genes from two chromosomes, and the mutation operation randomly alters the state of a gene. When the environment is changing, the population keeps adapting to the best fit point. When the environment is fixed, the population tends to converge to the sub-optimal point(s). By the assistance of the mutation, there are always evolved solutions that are deviated from the sub-optimal point. The diversifying process makes the population converge to the global-best solution eventually. The goal of the project “A Study of Image Processing Techniques with Genetic Programming” is to develop an image processing framework that incorporates genetic programming (and related evolutionary methods) as a tool. In this project, we will: (1) study the GP techniques and develop a method to reduce the GP complexity when processing image data; (2) the parallelism in image processing is potentially suitable to work with GP, and we will design the framework that can utilize the parallelism for enhancing the performance; (3) evaluate our design by comparing the performance produced by the conventional processing results. This project combines the image processing with GP techniques. The experiences and the results obtained during the study, design, and simulation would be very useful in the future related researches.
    Appears in Collections:[資訊創新與科技學系] 研究報告

    Files in This Item:

    There are no files associated with this item.

    All items in 機構典藏 are protected by copyright, with all rights reserved.


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback