English  |  正體中文  |  简体中文  |  Items with full text/Total items : 70588/70588 (100%)
Visitors : 23155371      Online Users : 728
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU 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://ir.lib.ncu.edu.tw/handle/987654321/71955

    Title: 台灣近海鮪魚之魚類辨識;Fish Recognition for Tunas in Adjacent Seas of Taiwan
    Authors: 張朝鈞;Chang,Chao-Chun
    Contributors: 通訊工程學系
    Keywords: 鮪魚;魚類辨識;部位切割;梯度向量直方圖;支持向量機;Tuna;fish recognition;part segmentation;histogram of oriented gradients;support vector machines
    Date: 2016-07-22
    Issue Date: 2016-10-13 14:08:55 (UTC+8)
    Publisher: 國立中央大學
    Abstract: 觀察國際間近年研究魚類主題以來,絕大部分都以辨識海洋景觀魚類為主要研究方向,較少研究是以辨識大型經濟魚類做為主題,因此本論文針對台灣近海之四種鮪魚,透過對鮪魚外觀特徵辨識,達成鮪魚和其它十九種非鮪魚的魚類辨識。
    本論文在訓練階段時將魚部位手動切割開並個別訓練其對應之分類器(classifier),並以各魚鰭部位之HOG descriptors個別訓練其對應之分類器(classifier),增加分類器對於局部影像特徵的辨識能力。辨識階段以本論文提出之二值投影法(binary projection)將魚類身體外觀三種特徵切割出,再透過Histogram of Oriented Gradients (HOG)將切割後的部位影像抓出其特徵描述子(descriptor),將其結果輸入至訓練階段所得之三種支持向量機分類器(support vector machine),在最後階段時,則整合三個(第一背鰭、第二背鰭、尾鰭)支持向量機分類結果透過投票(majority voting)機制決定是否為所要辨識的鮪魚種類。實驗結果顯示,四種鮪魚之平均辨識率約為72%,若只採用第二背鰭分類器,則辨識率可上升至80%。
    ;In recent years, most fish recognition algorithm focus on recognizing aquarium fish rather than large commercial fishes. Thus, this thesis focuses on 4 species of tuna in adjacent seas of Taiwan. Through recognizing features of appearance of tuna, the proposed scheme can differentiate 4 species of tuna from 19 species of non-tuna fish.
    At the training stage, this thesis segments fish fins manually and train SVM classifiers using the HOG descriptors. Based on the local image features, the method can improve recognition accuracy. In the test stage, this paper proposes to use binary projection for part segmentation. Descriptors of histogram of oriented gradients of three fins are the input of SVM classifiers and the classification results are majority voted for final decisions. Tests show that the recognition accuracy is around 72%. If the classification decision only depends on the feature of the second dorsal fin, the recognition accuracy is around 80%.
    Appears in Collections:[通訊工程研究所] 博碩士論文

    Files in This Item:

    File Description SizeFormat

    All items in NCUIR are protected by copyright, with all rights reserved.

    社群 sharing

    ::: Copyright National Central University. | 國立中央大學圖書館版權所有 | 收藏本站 | 設為首頁 | 最佳瀏覽畫面: 1024*768 | 建站日期:8-24-2009 :::
    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library IR team Copyright ©   - Feedback  - 隱私權政策聲明