Publications-Theses
Article View/Open
Publication Export
-
題名 圖像資料結構化與分類的探討
A Study of Image Structurization and Classification作者 莊于萱
Zhuang, Yu-Xuan貢獻者 余清祥<br>陳麗霞
莊于萱
Zhuang, Yu-Xuan關鍵詞 圖像辨識
資料結構化
圖像風格
分割圖像
解析度
Image Recognition
Data Structurization
Image Style
Splitting method
Resolution日期 2021 上傳時間 2-Sep-2021 15:45:13 (UTC+8) 摘要 資料以各種形態存在於我們生活中，寫過的每一篇文章，甚至拍過的每一張照片，透過適當的數位化皆可由量化分析挖掘出其中的重要訊息。過去資料分析大多侷限在數字格式，隨著電腦相關技術的發展，資訊解讀擴展至文字、圖像、音樂等各種類型的資料，我們的生活因為資訊傳遞快速、即時判讀而更加便利，影像辨識、自動駕駛等應用就是眾所周知的應用。資料格式多元、傳遞交換便捷，都是大數據時代的特點，使得資訊安全及品質愈形重要，如何解讀龐雜的大數據，更是政府及個人必備的關鍵能力。不具固定格式資料稱為非結構資料，而解讀這類型資料的首要挑戰為數位化格式，但轉檔方式與研究目標、資料屬性關係密切，很難訂出一個絕對標準。以圖像辨識為例，資料應轉換成三原色（紅綠藍，RGB：Red、Green、Blue）或是圖像形狀及大小，至今仍無定論；即便是以顏色紀錄，是否也需考量色彩飽和度、亮度等資訊？有鑑於圖像資料尚無統一的格式化，本文以視覺感受的方式定義變數，比較冷暖色、RGB、灰階、邊緣檢測、分割圖像等方法，協助分類不同風格的圖像。由於圖像辨識結果多半與其屬性有關（Data Dependent），本文分析三種類型的圖像資料：臺灣報紙頭版、美國Vogue雜誌封面、十九世紀油畫（現實派、印象派），其內容包含文字、照片（及圖片）、繪畫，再結合統計分析、機器學習模型，藉由電腦模擬與交叉驗證，探討不同的圖像格式方法、模型及其參數的分類準確性。研究結果顯示，分類準確性隨著解析度上升而提高，100100即有不錯的效果；而圖像格式化中以分割圖像法最佳，用於不同圖像資料及分析模型的分類準確性較高。
Data exists in various forms, including articles we write and photos we take. We can dig out important information through appropriate digitization and quantitative analysis. Data analysis was usually limited to data with digital format and now has expanded to all kinds of data, such as text, images, and music. Rapid data transmission and real-time analysis brings convenience to our lives, and image recognition and autonomous driving are two famous applications. The ability of analyzing messy and complicated data is a key ability in Big Data era. However, more than 90% of data are unstructured data and it needs to convert them into digital format before plugging into data analysis. But the conversion method is closely related to the research objectives and data attributes. Taking image recognition as an example, there is no consensus if the image data should be converted into three primary colors (red, green and blue, RGB) or their shape and size should also be considered.This study aims to explore different structurization methods for image data and evaluate which method, after plugging into classification models, has the highest accuracy in classifying images . We consider three types of image data: Taiwanese newspapers, Vogue magazine, and nineteenth century oil paintings, since the classification results are often data-dependent. We will apply statistical and machine learning models to explore classification accuracy of different image format methods. The analysis results show that the classification accuracy increases when the resolution becomes higher, and 100100 resolution can provide sufficiently satisfactory results. We found that the splitting method has highest accuracy in image classification, for three types of image data and different classification models.參考文獻 一、中文文獻孔萬增、朱善安（2007）。「基於切割子模塊的單樣本人臉識別」，《光電工程》，34（8），頁110-114。任大勇、賈振紅、楊傑（2019）。「結合位圖切割和區域合併的彩色圖像分割」，《計算機工程與應用》，55（2），頁162-167。柯裕嘉（2011）。「報紙消費者對頭版新聞形式與內容喜好度研究」，國立台灣師範大學圖文傳播學系碩士論文，頁1-120。胡毅（2015）。「米勒《拾穗》賞析」，《時代文學（下半月）》，第7期，頁75-75。施振祥（2010）。「基於顏色和邊緣信息分佈的圖像檢索」，《計算機科學》，37（2），頁256-260。辜衛東、李兵（2018）。「基於隨機區域合併的自動彩色圖像分割算法」，《計算機科學》，45（9），頁279-282。黃衍翠（2010）。「從《日出·印象》談印象派油畫之美」，《時代文學（上半月）》，第3期，頁229-231。楊賢藝（2006）。「論印象派繪畫的藝術特色」，《藝術教育》，第4期，頁94-95。劉振源（1900）。「印象派繪畫」，藝術圖書出版社，頁1-252。龔如森（2016）。「西班牙藝術夜空裡的星光-寫實主義的委拉斯蓋茲與浪漫主義的哥雅」，中國文化大學藝術學院美術學系學系碩士論文，頁1-136。二、英文文獻Barni, M., Pelagotti, A., and Piva, A.（2005）“Image processing for the analysis and conservation of paintings: opportunities and challenges.” IEEE Signal Process Magazine, 22, 141–144.Canny, J.（1986）“A computational approach to edge detection.” IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-8, 679–698.Cheng, Y.C. and Chen, S.Y.（2003）“Image classification using color, texture and regions. ” Image and Vision Computing, 21（9）, 759–776.Chen, W., Shi, Y.Q., and Xuan, G.（2007）“Identifying computer graphics using HSV color model and statistical moments of characteristic functions.” IEEE International Conference on Multimedia and Expo, 1123–1126.Christiana, W. （2014）“Female image in Vogue magazine: A pictorial analysis of facial and body language.” Robots Reading Vogue.Junhua, C. and Jing, L. （2012）“Research on color image classification based on HSV color space.” Second International Conference on Instrumentation, Measurement, Computer, Communication and Control, 944–947.Kabade, A.L. and Sangam, D.V.（2016） “Canny edge detection algorithm.” International Journal of Advanced Research in Electronics and Communication Engineering, 5（5）, 1292–1295.Kaur, B. and Garg, A. （2011）“Mathematical morphological edge detection for remote sensing images.” International Conference on Electronic Computer Technology, 5, 324–327.Lee, S.G. and Cha, E.Y. （2016）“Style classification and visualization of art painting’s genre using self-organizing maps.” Human-centric Computing and Information Sciences, 6, No.7.Rabby, M.K.M., Chowdhury, B., and Kim, J.H.（2018）“A modified canny edge detection algorithm for fruit detection and classification.” 10th International Conference on Electrical and Computer Engineering, 237–240.Pal, N.R. and Pal, S.K. （1993）“A review in image segmentation techniques.” Pattern Recognition, 26（9）, 1277–1294.Peter, L. （2013）“Vogue Cover Averages.” Robots Reading Vogue.Rong, W., Li, Z., Zhang, W. and Sun, L. （2014） “An improved canny edge detection algorithm.” Proceeding IEEE International Conference on Mechatronics and Automation, 2（2）, 577–582.Phoebe, P. （1997） “Impressionism.” Thames and Hudson, 1–187.Rohit, G. （2019）“LightGBM - Another gradient boosting algorithm.” , Retrieved March 28, 2019, from: https://rohitgr7.github.io/lightgbm-another-gradient-boosting/Süsstrunk, S., Buckley, R., and Swen, S. （1999）“Standard RGB color spaces.” Color and Imaging Conference, 127–134.Seetharaman, K. （2019） “Melanoma Image Classification Based on Color, Shape, and Texture Features Using Multivariate Statistical Tests.” Journal of Computational and Theoretical Nanoscience, 16（4）, 1717–1724.Song, M. and Civco, D. （2004） “Road extraction using SVM and image segmentation.” Photogrammetric Engineering and Remote Sensing, 70（12）, 1365–1371.Salman, A., Semwal, A., Bhatt, U., and Thakkar, V. M. （2017） “Leaf classification and identification using Canny Edge Detector and SVM classifier.” Proceedings of the 2017 International Conference on Inventive Systems and Control（ICISC）, Coimbatore, India, 19–20 January 2017, 1–4.Xin, M. and Wang, Y. （2019）“Research on image classification model based on deep convolution neural network.” EURASIP Journal on Image and Video Processing, 40.Denny, B. （2015）“Understanding Convolutional Neural Networks for NLP.” Retrieved November 7, 2015, from: http://www.wildml.com/2015/11/understanding-convolutional-neural-networks-for-nlp/ 描述 碩士
國立政治大學
統計學系
108354028資料來源 http://thesis.lib.nccu.edu.tw/record/#G0108354028 資料類型 thesis dc.contributor.advisor 余清祥<br>陳麗霞 zh_TW dc.contributor.author (Authors) 莊于萱 zh_TW dc.contributor.author (Authors) Zhuang, Yu-Xuan en_US dc.creator (作者) 莊于萱 zh_TW dc.creator (作者) Zhuang, Yu-Xuan en_US dc.date (日期) 2021 en_US dc.date.accessioned 2-Sep-2021 15:45:13 (UTC+8) - dc.date.available 2-Sep-2021 15:45:13 (UTC+8) - dc.date.issued (上傳時間) 2-Sep-2021 15:45:13 (UTC+8) - dc.identifier (Other Identifiers) G0108354028 en_US dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/136834 - dc.description (描述) 碩士 zh_TW dc.description (描述) 國立政治大學 zh_TW dc.description (描述) 統計學系 zh_TW dc.description (描述) 108354028 zh_TW dc.description.abstract (摘要) 資料以各種形態存在於我們生活中，寫過的每一篇文章，甚至拍過的每一張照片，透過適當的數位化皆可由量化分析挖掘出其中的重要訊息。過去資料分析大多侷限在數字格式，隨著電腦相關技術的發展，資訊解讀擴展至文字、圖像、音樂等各種類型的資料，我們的生活因為資訊傳遞快速、即時判讀而更加便利，影像辨識、自動駕駛等應用就是眾所周知的應用。資料格式多元、傳遞交換便捷，都是大數據時代的特點，使得資訊安全及品質愈形重要，如何解讀龐雜的大數據，更是政府及個人必備的關鍵能力。不具固定格式資料稱為非結構資料，而解讀這類型資料的首要挑戰為數位化格式，但轉檔方式與研究目標、資料屬性關係密切，很難訂出一個絕對標準。以圖像辨識為例，資料應轉換成三原色（紅綠藍，RGB：Red、Green、Blue）或是圖像形狀及大小，至今仍無定論；即便是以顏色紀錄，是否也需考量色彩飽和度、亮度等資訊？有鑑於圖像資料尚無統一的格式化，本文以視覺感受的方式定義變數，比較冷暖色、RGB、灰階、邊緣檢測、分割圖像等方法，協助分類不同風格的圖像。由於圖像辨識結果多半與其屬性有關（Data Dependent），本文分析三種類型的圖像資料：臺灣報紙頭版、美國Vogue雜誌封面、十九世紀油畫（現實派、印象派），其內容包含文字、照片（及圖片）、繪畫，再結合統計分析、機器學習模型，藉由電腦模擬與交叉驗證，探討不同的圖像格式方法、模型及其參數的分類準確性。研究結果顯示，分類準確性隨著解析度上升而提高，100100即有不錯的效果；而圖像格式化中以分割圖像法最佳，用於不同圖像資料及分析模型的分類準確性較高。 zh_TW dc.description.abstract (摘要) Data exists in various forms, including articles we write and photos we take. We can dig out important information through appropriate digitization and quantitative analysis. Data analysis was usually limited to data with digital format and now has expanded to all kinds of data, such as text, images, and music. Rapid data transmission and real-time analysis brings convenience to our lives, and image recognition and autonomous driving are two famous applications. The ability of analyzing messy and complicated data is a key ability in Big Data era. However, more than 90% of data are unstructured data and it needs to convert them into digital format before plugging into data analysis. But the conversion method is closely related to the research objectives and data attributes. Taking image recognition as an example, there is no consensus if the image data should be converted into three primary colors (red, green and blue, RGB) or their shape and size should also be considered.This study aims to explore different structurization methods for image data and evaluate which method, after plugging into classification models, has the highest accuracy in classifying images . We consider three types of image data: Taiwanese newspapers, Vogue magazine, and nineteenth century oil paintings, since the classification results are often data-dependent. We will apply statistical and machine learning models to explore classification accuracy of different image format methods. The analysis results show that the classification accuracy increases when the resolution becomes higher, and 100100 resolution can provide sufficiently satisfactory results. We found that the splitting method has highest accuracy in image classification, for three types of image data and different classification models. en_US dc.description.tableofcontents 第一章 緒論 1第一節 研究動機 1第二節 研究目的 2第二章 文獻探討 3第一節 文獻回顧 3第二節 資料介紹 5第三章 研究方法 9第一節 色彩空間 10第二節 Canny Algorithm 13第三節 顏色檢測 16第四節 分類模型 20第四章 圖像分析 25第一節 探索性資料分析 25第二節 參數設定 35第三節 圖像風格分類-二分類 37第四節 圖像風格分類-多分類 40第五章 結論與建議 44第一節 結論 44第二節 未來建議 45參考文獻 47 zh_TW dc.format.extent 2987737 bytes - dc.format.mimetype application/pdf - dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0108354028 en_US dc.subject (關鍵詞) 圖像辨識 zh_TW dc.subject (關鍵詞) 資料結構化 zh_TW dc.subject (關鍵詞) 圖像風格 zh_TW dc.subject (關鍵詞) 分割圖像 zh_TW dc.subject (關鍵詞) 解析度 zh_TW dc.subject (關鍵詞) Image Recognition en_US dc.subject (關鍵詞) Data Structurization en_US dc.subject (關鍵詞) Image Style en_US dc.subject (關鍵詞) Splitting method en_US dc.subject (關鍵詞) Resolution en_US dc.title (題名) 圖像資料結構化與分類的探討 zh_TW dc.title (題名) A Study of Image Structurization and Classification en_US dc.type (資料類型) thesis en_US dc.relation.reference (參考文獻) 一、中文文獻孔萬增、朱善安（2007）。「基於切割子模塊的單樣本人臉識別」，《光電工程》，34（8），頁110-114。任大勇、賈振紅、楊傑（2019）。「結合位圖切割和區域合併的彩色圖像分割」，《計算機工程與應用》，55（2），頁162-167。柯裕嘉（2011）。「報紙消費者對頭版新聞形式與內容喜好度研究」，國立台灣師範大學圖文傳播學系碩士論文，頁1-120。胡毅（2015）。「米勒《拾穗》賞析」，《時代文學（下半月）》，第7期，頁75-75。施振祥（2010）。「基於顏色和邊緣信息分佈的圖像檢索」，《計算機科學》，37（2），頁256-260。辜衛東、李兵（2018）。「基於隨機區域合併的自動彩色圖像分割算法」，《計算機科學》，45（9），頁279-282。黃衍翠（2010）。「從《日出·印象》談印象派油畫之美」，《時代文學（上半月）》，第3期，頁229-231。楊賢藝（2006）。「論印象派繪畫的藝術特色」，《藝術教育》，第4期，頁94-95。劉振源（1900）。「印象派繪畫」，藝術圖書出版社，頁1-252。龔如森（2016）。「西班牙藝術夜空裡的星光-寫實主義的委拉斯蓋茲與浪漫主義的哥雅」，中國文化大學藝術學院美術學系學系碩士論文，頁1-136。二、英文文獻Barni, M., Pelagotti, A., and Piva, A.（2005）“Image processing for the analysis and conservation of paintings: opportunities and challenges.” IEEE Signal Process Magazine, 22, 141–144.Canny, J.（1986）“A computational approach to edge detection.” IEEE Transactions on Pattern Analysis and Machine Intelligence, PAMI-8, 679–698.Cheng, Y.C. and Chen, S.Y.（2003）“Image classification using color, texture and regions. ” Image and Vision Computing, 21（9）, 759–776.Chen, W., Shi, Y.Q., and Xuan, G.（2007）“Identifying computer graphics using HSV color model and statistical moments of characteristic functions.” IEEE International Conference on Multimedia and Expo, 1123–1126.Christiana, W. （2014）“Female image in Vogue magazine: A pictorial analysis of facial and body language.” Robots Reading Vogue.Junhua, C. and Jing, L. （2012）“Research on color image classification based on HSV color space.” Second International Conference on Instrumentation, Measurement, Computer, Communication and Control, 944–947.Kabade, A.L. and Sangam, D.V.（2016） “Canny edge detection algorithm.” International Journal of Advanced Research in Electronics and Communication Engineering, 5（5）, 1292–1295.Kaur, B. and Garg, A. （2011）“Mathematical morphological edge detection for remote sensing images.” International Conference on Electronic Computer Technology, 5, 324–327.Lee, S.G. and Cha, E.Y. （2016）“Style classification and visualization of art painting’s genre using self-organizing maps.” Human-centric Computing and Information Sciences, 6, No.7.Rabby, M.K.M., Chowdhury, B., and Kim, J.H.（2018）“A modified canny edge detection algorithm for fruit detection and classification.” 10th International Conference on Electrical and Computer Engineering, 237–240.Pal, N.R. and Pal, S.K. （1993）“A review in image segmentation techniques.” Pattern Recognition, 26（9）, 1277–1294.Peter, L. （2013）“Vogue Cover Averages.” Robots Reading Vogue.Rong, W., Li, Z., Zhang, W. and Sun, L. （2014） “An improved canny edge detection algorithm.” Proceeding IEEE International Conference on Mechatronics and Automation, 2（2）, 577–582.Phoebe, P. （1997） “Impressionism.” Thames and Hudson, 1–187.Rohit, G. （2019）“LightGBM - Another gradient boosting algorithm.” , Retrieved March 28, 2019, from: https://rohitgr7.github.io/lightgbm-another-gradient-boosting/Süsstrunk, S., Buckley, R., and Swen, S. （1999）“Standard RGB color spaces.” Color and Imaging Conference, 127–134.Seetharaman, K. （2019） “Melanoma Image Classification Based on Color, Shape, and Texture Features Using Multivariate Statistical Tests.” Journal of Computational and Theoretical Nanoscience, 16（4）, 1717–1724.Song, M. and Civco, D. （2004） “Road extraction using SVM and image segmentation.” Photogrammetric Engineering and Remote Sensing, 70（12）, 1365–1371.Salman, A., Semwal, A., Bhatt, U., and Thakkar, V. M. （2017） “Leaf classification and identification using Canny Edge Detector and SVM classifier.” Proceedings of the 2017 International Conference on Inventive Systems and Control（ICISC）, Coimbatore, India, 19–20 January 2017, 1–4.Xin, M. and Wang, Y. （2019）“Research on image classification model based on deep convolution neural network.” EURASIP Journal on Image and Video Processing, 40.Denny, B. （2015）“Understanding Convolutional Neural Networks for NLP.” Retrieved November 7, 2015, from: http://www.wildml.com/2015/11/understanding-convolutional-neural-networks-for-nlp/ zh_TW dc.identifier.doi (DOI) 10.6814/NCCU202101197 en_US
