1. NCCU Academic Hub
  2. 學術產出
  3. College of Informatics
  4. Theses
  5. Item

學術產出-Theses

Article View/Open

Publication Export

Google ScholarTM

政大圖書館

Citation Infomation

  • No doi shows Citation Infomation
題名 Instagram相片之色彩分析及應用
Color analysis of Instagram photos and its application
作者 林儀婷
Lin, Yi-Ting
貢獻者 廖文宏<br>廖文宏
Liao, Wen-Hung
林儀婷
Lin, Yi-Ting
關鍵詞 社群網絡
色彩空間分析
機器學習
情感分析
形象和情感
Social network
Color space analysis
Machine learning
Sentiment analysis
Image and emotion
日期 2016
上傳時間 8-Feb-2017 16:51:04 (UTC+8)
摘要 近來Instagram成為流行的分享照片社交平台。在上傳影像到網路社交平台時，人們透過套用不同的濾鏡來表達他們的感受。然而，對於修改過的影像，我們不太可能逆向推回得知影像套用了什麼樣的濾鏡。本研究嘗試透過定義出十種影像風格，對應於一些最常應用的濾鏡，來解決這種逆向工程問題。因此，原始問題被轉化為分類問題，並可以使用機器學習方法來解決。為了生成訓練數據，我們根據用戶投票收集標記的結果。根據我們的實驗，在調查中概述的十個類別中，投票的結果有很高的共識。我們在HSV空間中使用分析出的顏色特徵來區分影像風格，並採用支持向量機（SVM）做分類。驗證我們數據集中的Top 1和Top 3準確度分別為64%和96%，顯示機器分類的效能與人類觀察者的效能相當。最後，我們導入數位著名攝影師的作品，進行個案研究，以測試風格識別和情感分析結果。
Recently, Instagram has become a very popular social media platform for sharing photos. People apply different type of filters to express their feelings when posting photos on social networking sites. Given a filtered image, it is difficult, if not possible, to determine which filter has been applied to obtain the observed effects. This study attempts to address this reverse engineering problem by defining ten image styles corresponding to some of the most frequently applied filters. As such, the original question is cast into a classification problem which can be solved using machine learning approaches. To generate training data, we collected the labeled results based on user votes. Consensuses among users are found to be high in the ten categories outlined in our investigation. We employ color features in the HSV space to characterize image styles. Support vector machine (SVM) is then used for classification. The accuracies for top-1 and top-3 category using our dataset are 64% and 96%, respectively. The performance of machine classification is comparable to that of human observers. Finally, works by famous photographers are brought in to validate the style recognition and sentiment analysis results.
參考文獻 [1] 世界衛生組織統計之憂鬱症人數http://www.who.int/mediacentre/factsheets/fs369/en/
[2] Gonzalez, Rafael C., and E. Richard. "Woods, digital image processing." ed: Prentice Hall Press, ISBN 0-201-18075-8 (2002).
[3] Wikipedia contributors, "HSL and HSV," Wikipedia, The Free Encyclopedia, https://en.wikipedia.org/w/index.php?title=HSL_and_HSV&oldid=756001150 (accessed December 21, 2016).
[4] 色彩心理學：http://td026544.pixnet.net/blog/post/32030867-%E8%89%B2%E5%BD%A9%E5%BF%83%E7%90%86%E5%AD%B8
[5] 彭姝樺 色彩暗號：關於那些人的顏色學，心理事。尖端出版股份有限公司，2011。
[6] Reece, Andrew G., and Christopher M. Danforth. "Instagram photos reveal predictive markers of depression." arXiv preprint arXiv:1608.03282 (2016).
[7] Lee, Honglak, et al. "Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations." Proceedings of the 26th annual international conference on machine learning. ACM, 2009.
[8] Shin, Laura (June 26, 2012). "Google brain simulator teaches itself to recognize cats". SmartPlanet. Retrieved February 11, 2014.
[9] Cortes, C.; Vapnik, V. "Support-vector networks". Machine Learning. 1995, 20 (3): 273–297.
[10] Hsu, Chih-Wei, Chih-Chung Chang, and Chih-Jen Lin. "A practical guide to support vector classification." (2003): 1-16.
[11] LIBSVM https://www.csie.ntu.edu.tw/~cjlin/libsvm/
[12] Pang, Bo, and Lillian Lee. "Opinion mining and sentiment analysis." Foundations and trends in information retrieval 2.1-2 (2008): 1-135.
[13] 彭聲揚 透過圖片標籤觀察情緒字詞與事物概念之關聯，政治大學碩士論文，2011年7月
[14] 陳育修 藉由孿生網路進行不受濾鏡影響之社群網路圖片分類，台灣大學碩士碩文，2015年7月
[15] Barrett, Lisa Feldman. "Valence is a basic building block of emotional life." Journal of Research in Personality 40.1 (2006): 35-55.
描述 碩士
國立政治大學
資訊科學系碩士在職專班
103971007
資料來源 http://thesis.lib.nccu.edu.tw/record/#G0103971007
資料類型 thesis
dc.contributor.advisor 廖文宏<br>廖文宏zh_TW
dc.contributor.advisor Liao, Wen-Hungen_US
dc.contributor.author (Authors) 林儀婷zh_TW
dc.contributor.author (Authors) Lin, Yi-Tingen_US
dc.creator (作者) 林儀婷zh_TW
dc.creator (作者) Lin, Yi-Tingen_US
dc.date (日期) 2016en_US
dc.date.accessioned 8-Feb-2017 16:51:04 (UTC+8)-
dc.date.available 8-Feb-2017 16:51:04 (UTC+8)-
dc.date.issued (上傳時間) 8-Feb-2017 16:51:04 (UTC+8)-
dc.identifier (Other Identifiers) G0103971007en_US
dc.identifier.uri (URI) http://nccur.lib.nccu.edu.tw/handle/140.119/106477-
dc.description (描述) 碩士zh_TW
dc.description (描述) 國立政治大學zh_TW
dc.description (描述) 資訊科學系碩士在職專班zh_TW
dc.description (描述) 103971007zh_TW
dc.description.abstract (摘要) 近來Instagram成為流行的分享照片社交平台。在上傳影像到網路社交平台時，人們透過套用不同的濾鏡來表達他們的感受。然而，對於修改過的影像，我們不太可能逆向推回得知影像套用了什麼樣的濾鏡。本研究嘗試透過定義出十種影像風格，對應於一些最常應用的濾鏡，來解決這種逆向工程問題。因此，原始問題被轉化為分類問題，並可以使用機器學習方法來解決。為了生成訓練數據，我們根據用戶投票收集標記的結果。根據我們的實驗，在調查中概述的十個類別中，投票的結果有很高的共識。我們在HSV空間中使用分析出的顏色特徵來區分影像風格，並採用支持向量機（SVM）做分類。驗證我們數據集中的Top 1和Top 3準確度分別為64%和96%，顯示機器分類的效能與人類觀察者的效能相當。最後，我們導入數位著名攝影師的作品，進行個案研究，以測試風格識別和情感分析結果。zh_TW
dc.description.abstract (摘要) Recently, Instagram has become a very popular social media platform for sharing photos. People apply different type of filters to express their feelings when posting photos on social networking sites. Given a filtered image, it is difficult, if not possible, to determine which filter has been applied to obtain the observed effects. This study attempts to address this reverse engineering problem by defining ten image styles corresponding to some of the most frequently applied filters. As such, the original question is cast into a classification problem which can be solved using machine learning approaches. To generate training data, we collected the labeled results based on user votes. Consensuses among users are found to be high in the ten categories outlined in our investigation. We employ color features in the HSV space to characterize image styles. Support vector machine (SVM) is then used for classification. The accuracies for top-1 and top-3 category using our dataset are 64% and 96%, respectively. The performance of machine classification is comparable to that of human observers. Finally, works by famous photographers are brought in to validate the style recognition and sentiment analysis results.en_US
dc.description.tableofcontents 第一章 緒論 11
1.1 研究背景 12
1.2 感覺及情感研究之重要性 12
1.3 色彩分析之應用 13
1.4 研究動機與目的 13
1.5 研究架構 14
第二章 領域知識與相關研究 15
2.1 色彩分析 15
2.1.1 HSL和HSV色彩空間 15
2.1.3 選擇HSV色彩空間 15
2.1.4 色彩心理學 16
2.1.5 色彩心理學-色彩感情 19
2.2 濾鏡 19
2.3 機器學習 23
2.3.1 深度學習 23
2.3.2 SVM (Support Vector Machines) 24
2.4 情感分析（Sentiment Analysis） 26
2.4.1 維度模式情緒表示法 26
第三章 資料分析與模型建立 29
3.1 研究架構 29
3.1.1 資料來源 29
3.1.1.1 Instagram攝影社群 29
3.1.1.2 定義風格類別 30
3.1.2 資料前處理 36
3.1.2.1 影像前測、投票 36
3.1.2.2 HSV色彩分析 40
3.2 使用SVM 訓練資料並預測 43
3.3 分析結果 44
3.4 搭配情感分析來驗證影像 45
3.4.1 情感分析之應用 45
3.5 小結 46
第四章 實驗結果與評估 47
4.1 挑選訓練資料樣本影像 47
4.2 SVM 訓練及預測結果 47
4.2.1 預測錯誤的影像分析 53
4.3 影像色彩分析搭配情感分析之應用 56
4.3.1 Instagram攝影師之影像分析 56
4.4 小結 71
第五章 結論與未來研究 73
5.1 研究成果 73
5.1.1 機器能夠辨別出影像風格的差異 73
5.1.2 分析易混淆之風格 73
5.1.3 影像帶給人的感覺與情感分析有相關 74
5.2 未來研究 74
5.2.1 影像內容的分析 74
5.2.2 影像風格的延伸 75
5.2.3 情緒分析的應用 75		zh_TW
dc.format.extent 6410292 bytes-
dc.format.mimetype application/pdf-
dc.source.uri (資料來源) http://thesis.lib.nccu.edu.tw/record/#G0103971007en_US
dc.subject (關鍵詞) 社群網絡zh_TW
dc.subject (關鍵詞) 色彩空間分析zh_TW
dc.subject (關鍵詞) 機器學習zh_TW
dc.subject (關鍵詞) 情感分析zh_TW
dc.subject (關鍵詞) 形象和情感zh_TW
dc.subject (關鍵詞) Social networken_US
dc.subject (關鍵詞) Color space analysisen_US
dc.subject (關鍵詞) Machine learningen_US
dc.subject (關鍵詞) Sentiment analysisen_US
dc.subject (關鍵詞) Image and emotionen_US
dc.title (題名) Instagram相片之色彩分析及應用zh_TW
dc.title (題名) Color analysis of Instagram photos and its applicationen_US
dc.type (資料類型) thesisen_US
dc.relation.reference (參考文獻) [1] 世界衛生組織統計之憂鬱症人數http://www.who.int/mediacentre/factsheets/fs369/en/
[2] Gonzalez, Rafael C., and E. Richard. "Woods, digital image processing." ed: Prentice Hall Press, ISBN 0-201-18075-8 (2002).
[3] Wikipedia contributors, "HSL and HSV," Wikipedia, The Free Encyclopedia, https://en.wikipedia.org/w/index.php?title=HSL_and_HSV&oldid=756001150 (accessed December 21, 2016).
[4] 色彩心理學：http://td026544.pixnet.net/blog/post/32030867-%E8%89%B2%E5%BD%A9%E5%BF%83%E7%90%86%E5%AD%B8
[5] 彭姝樺 色彩暗號：關於那些人的顏色學，心理事。尖端出版股份有限公司，2011。
[6] Reece, Andrew G., and Christopher M. Danforth. "Instagram photos reveal predictive markers of depression." arXiv preprint arXiv:1608.03282 (2016).
[7] Lee, Honglak, et al. "Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations." Proceedings of the 26th annual international conference on machine learning. ACM, 2009.
[8] Shin, Laura (June 26, 2012). "Google brain simulator teaches itself to recognize cats". SmartPlanet. Retrieved February 11, 2014.
[9] Cortes, C.; Vapnik, V. "Support-vector networks". Machine Learning. 1995, 20 (3): 273–297.
[10] Hsu, Chih-Wei, Chih-Chung Chang, and Chih-Jen Lin. "A practical guide to support vector classification." (2003): 1-16.
[11] LIBSVM https://www.csie.ntu.edu.tw/~cjlin/libsvm/
[12] Pang, Bo, and Lillian Lee. "Opinion mining and sentiment analysis." Foundations and trends in information retrieval 2.1-2 (2008): 1-135.
[13] 彭聲揚 透過圖片標籤觀察情緒字詞與事物概念之關聯，政治大學碩士論文，2011年7月
[14] 陳育修 藉由孿生網路進行不受濾鏡影響之社群網路圖片分類，台灣大學碩士碩文，2015年7月
[15] Barrett, Lisa Feldman. "Valence is a basic building block of emotional life." Journal of Research in Personality 40.1 (2006): 35-55.		zh_TW