跳到主要內容

臺灣博碩士論文加值系統

(35.172.136.29) 您好!臺灣時間:2021/07/29 08:36
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

我願授權國圖
: 
twitterline
研究生:李諺宗
研究生(外文):Yan-Zong Li
論文名稱:以主成分分析及模糊方法判斷臉部膚質
論文名稱(外文):Determining Facial Skin Quality by Principal Component Analysis and Fuzzy Model
指導教授:黃有評黃有評引用關係
口試委員:謝尚琳朱鴻棋黃正民
口試日期:2012-07-11
學位類別:碩士
校院名稱:國立臺北科技大學
系所名稱:電機工程系所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2012
畢業學年度:100
語文別:中文
論文頁數:66
中文關鍵詞:模糊方法主成分分析灰階共生矩陣哈爾小波轉換智慧型手機
外文關鍵詞:Fuzzy modelprincipal component analysisgray level co-occurrence matrixHaar wavelet transformsmart phones
相關次數:
  • 被引用被引用:1
  • 點閱點閱:251
  • 評分評分:
  • 下載下載:0
  • 收藏至我的研究室書目清單書目收藏:0
各種不同的膚質檢測儀隨著科技的發展應運而生,其中包含大型膚質檢測儀及小型膚質檢測儀。大型膚質檢測儀藉由影像處理的方式可詳細的檢測出皮膚的狀況及種類,包含紋理、水分、油分、柔軟度等特性,但相對的價格較昂貴;而小型膚質檢測儀攜帶方便且輕巧,可簡單檢測出皮膚的水分、油分及柔軟度,雖然價格便宜,但需透過檢測儀前端的金屬以接觸式的方式量測。隨著智慧型手機攝影功能的增強,數位影像的拍攝與取得變得容易。因此本研究使用智慧型手機拍攝臉部皮膚影像,藉由簡單拍照的方式進行影像特徵的擷取,將取得的臉部皮膚影像轉成灰階值,再透過灰階共生矩陣計算紋理特徵,包含對比度、均勻度及同質度,接著對原始灰階影像使用哈爾小波轉換獲得垂直、水平及對角紋理的紋理特徵。定義六種紋理特徵之輸入歸屬函數及膚質種類之輸出歸屬函數,藉由所設計的模糊規則庫推論出對應的膚質特性。為了減少膚質判斷的特徵與運算時間量,本研究使用主成分分析的方法篩選出較具鑑別度之紋理特徵,藉由較少的特徵推論出對應的膚質種類。以六種紋理特徵經模糊推論結果當作比較參考值,則以主成分分析之紋理特徵與灰階共生矩陣所得紋理特徵的兩種結果作誤差分析的比較,準確性分別為96.29%及93.21%,此結果驗證主成分分析較具膚質判斷的效果。

With the development of science and technology, a variety of skin detectors emerged, including the large ones and the small ones. Large skin detectors can give information in detail by image processing, which includes the condition and the type of skin. We can further divide them into texture, moisture, oil, softness and other characteristics, but the cost of the detectors is much higher than small skin detectors that are easy to carry due to their lightweight and easy to use. But the small ones need to contact the face by the metal part of the detector while measuring. As photographic enhancements of the smart phones, taking and acquiring digital images become easier. Thus, this study uses smart phone to take facial skin images. The image features are extracted from digital images and converted into gray level. Then, we can calculate the texture features, including contrast, entropy and inverse difference moment through gray level co-occurrence matrix. Finally vertical, horizontal and diagonal texture features on the original gray image were found by Haar wavelet transform. After defining the six texture features of input and output membership functions of the skin types, the skin quality characteristics are inferred by the proposed the fuzzy models. In order to reduce the computing time, we use principal component analysis method to discriminate texture features. The purpose is to find the corresponding skin types with fewer features. With the six texture features from fuzzy inference results as a reference value, both the results from the principal component analysis with texture features and gray level co-occurrence matrix with texture features have the accuracy rates of 96.29% and 93.21%, respectively. These results verify that the principal components analysis can provide better detection accuracy.

摘 要 i
ABSTRACT ii
誌 謝 iv
目 錄 v
表目錄 vii
圖目錄 viii
第一章 緒論 1
1.1 研究背景 1
1.2 研究目的 2
1.3 研究方法 2
1.4 論文架構 3
第二章 相關技術與運用探討 4
2.1 紋理特徵 4
2.1.1 灰階共生矩陣 4
2.1.2 哈爾小波轉換 8
2.2 主成分分析 15
2.3 模糊理論 18
2.3.1 模糊集合 19
2.3.2 模糊歸屬函數 20
2.3.3 模糊運算 23
2.4 Android作業系統 23
2.4.1 系統架構 24
2.4.2 生命週期 27
第三章 系統架構與設計 30
3.1 系統架構 30
3.2 系統流程與設計 35
3.2.1 系統執行流程 35
3.2.2膚質檢測模糊系統 36
3.2.2.1 GLCM膚質檢測模糊系統設計 37
3.2.2.2 結合GLCM及Haar Wavelet Transform膚質檢測模糊系統計 40
3.2.2.3 PCA膚質檢測模糊系統設計 44
3.3 開發環境 47
3.3.1 硬體規格 47
3.3.2 軟體版本 48
第四章 實驗結果與分析 49
4.1 膚質分析與結果 49
4.2 系統介面 57
第五章 結論與未來展望 62
5.1 結論 62
5.2 未來展望 63
參考文獻 64


[1]維基百科 – Natural skin care,
http://en.wikipedia.org/wiki/Natural_skin_care。
[2]J.-D. Lee, Y.-C. Chen, L.-C. Liu and C.-H. Huang, “A facial-skin condition classification system using texture and fuzzy c-means technologies,” in Proc. of IEEE Int. Conf. on Fuzzy Systems and Knowledge Discovery, Taoyuan, Taiwan, pp. 24-27, Aug. 2007.
[3]A. L. Amet, A. Ertuzun, and A. Ercil, “Texture defect detection using subband domain co-occurrence matrices,” in Proc. of IEEE Int. Conf. on Image Analysis and Interpretation, Istanbul, Turkey, pp. 205-210, Apr. 1998.
[4]L. H. Siew, R. M. Hodgson, and E. J. Wood, “Texture measures for carpet wear assessment,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 10, no. 1, pp.92-105, 1988.
[5]D. Mery, M. A. Berti “Automatic detection of welding defects using texture features,” in Proc. of Symp. on Computerized Tomography and Image Processing for Industrial Applications in Radiology , Berlin, German, pp.23-25, June 2003.
[6]Rafael C. Gonzalez, Richard E. Woods, “Digital Image Processing”, Prentice-Hall International Inc., Taipei, Taiwan, 1998.
[7]鐘國亮,影像處理與電腦視覺(第四版),臺北:東華書局,2008。
[8]鐘國亮,資料壓縮的原理與應用(第二版),臺北:全華科技圖書,2004。
[9]R. M. Haralick, K. Shanmugam, and I. H. Dinstein, “Textural features for image classification,” IEEE Trans. on Systems, Man and Cybernetics, vol. SMC-3, no. 3, pp.610-621, 1973.
[10]R. M. Haralick, “Statistical and structural approaches to texture,” Proceedings of the IEEE, vol. 67, no. 5, pp.786-804, 1979.
[11]C. Chuan-Yu, L. Chi-Fang, and W. Hung-Jen, “Using principal component analysis and radial basis function neural networks for image contents classification,” in Proc. of IEEE Int. Conf. on Intelligent Information Hiding and Multimedia Signal Processing, Yunlin, Taiwan, pp.7-10, Nov. 2007.
[12]維基百科 – 哈爾小波轉換,
http://zh.wikipedia.org/zh-tw/%E5%93%88%E7%88%BE%E5%B0%8F%E6%B3%A2%E8%BD%89%E6%8F%9B。
[13]V. Balamurugan and P. AnandhaKumar, “An integrated color and texture feature based framework for content based image retrieval using 2D Wavelet Transform,” in Proc. of IEEE Int. Conf. on Computing, Communication and Networking, Chennai, India, pp.1-16, Dec. 2008.
[14]Y. Yang, Z. Su, and L. Sun, “Medical image enhancement algorithm based on wavelet transform,” Electronics Letters, vol. 46, no. 2, pp.120-121, 2010.
[15]維基百科 – 主成分分析,
http://zh.wikipedia.org/zh-tw/%E4%B8%BB%E6%88%90%E5%88%86%E5%88%86%E6%9E%90。
[16]L. Zhao and Z. Guo, “Face recognition method based on adaptively weighted block-two dimensional principal component analysis,” in Proc. of IEEE Int. Conf. on Computational Intelligence, Communication Systems and Networks (CICSyN), Shenyang , China, pp.22-25, July 2011.
[17]H. Gyeongyong, P. Gader, and H. Frigui, “Robust kernel PCA using fuzzy membership,” in Proc. of IEEE Int. Conf. on Neural Networks, Gainesville, FL, USA, pp.1213-1220, June 2009.
[18]Y. Jian, D. Zhang, and Y. Jing-Yu, “Constructing PCA baseline algorithms to reevaluate ICA-Based face-recognition performance,” IEEE Tran. on Systems, Man, and Cybernetics, Part B: Cybernetics, vol. 37, no. 4, pp.1015-1021, 2007.
[19]Steven J. Leon, 線性代數(第七版),臺北:高立圖書,2008。
[20]汪惠健譯,模糊理論與應用,臺北:臺灣培生教育出版公司,2006。
[21]A. Singh, C. Quek, S. -Y. Cho, “DCT-Yager FNN: A novel Yager-based fuzzy neural network with the discrete clustering technique,” IEEE Trans. on Neural Networks, vol. 19, no. 4, pp.625-644, Feb. 2008.
[22]Open Handset Alliance, http://www.openhandsetalliance.com。
[23]維基百科 – Android, http://zh.wikipedia.org/wiki/Android。
[24]What is Android, http://developer.android.com/guide/basics/what-is-android.html。
[25]Activity, http://developer.android.com/reference/android/app/Activity.html。
[26]HTC 宏達電,http://www.htc.com/tw/specification.aspx?p_id=283。


QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top