臺灣博碩士論文加值系統

皮革是鞋具用品中，使用最多的一種材料，皮革在製成後，具有獨特的天然紋理和觸感，可增加產品的質感。本研究以感性工學為基礎，分析鞋具皮革顏色、紋理等特徵，结合類神經網路驗證，將消費者的視覺與視觸覺意象，反應在皮革類型上。提供給鞋類設計者與皮料開發廠商選用與設計之建議參考，加速設計流程並輔助設計師以客觀的方式，進行鞋具皮革選材，提升鞋具產品創新的競爭力。
研究首先列出消費者對鞋具皮革在視覺與視觸覺感性知覺的代表詞彙對，藉由製鞋廠商提供的鞋具皮革樣片54片，進行視覺與視觸覺感性問卷調查評分後，將各皮革的視覺、視觸覺感性量化編譯。再通过撰寫程式，擷取皮革樣片影像色彩的主色與黏著度作為色彩特徵；分析影像灰階值，以像素八鄰域為基，提出LBP, SCOV, VAR, SAC等相關計算方法，進行影像紋理特徵的擷取。將獲取的色彩特徵值與紋理特徵值作為輸入層，感性語彙量化值作為輸出層，以49個皮革樣片進行倒傳遞類神經網路訓練，5個皮革樣片進行測試，完成3類17種組合型態之類神經網路訓練驗證，並獲得最佳之結果為：色彩特徵搭配VAR紋理特徵為輸入層，視觸覺感性量化值作為輸出層的倒傳遞類神經網路。
本研究具體成果如下：（1）提出鞋具皮革色彩特徵和紋理特徵的參數轉換方法；（2）得出較佳的鞋具皮革感性工學與類神經網路訓練方法，建立以感性語彙為基礎，結合倒傳遞類神經網路的鞋具皮革特徵輔助設計流程；（3）設計電腦輔助鞋具皮革評價查詢系統，設計師可藉以找出最接近建議的皮革樣片。

As the most widely used material in footwear, leather has its unique texture and hand feeling which could improve the quality of the products. Based on Kansei Engineering, this study analyzed the features of shoe leather, including colors and textures, and then expresses the vision and visual-tactile imagery of the consumers on the types of leather through artificial neural network verification. This paper also provided suggestions for footwear designers and the leather manufacturers on the design and selection of leather to accelerate the design flow. To assist the designers with the selection of suitable materials in an objective way, it promoted the innovative competitiveness of the footwear eventually.
This study firstly listed the representative words of the consumers on vision and visual-tactile sense perception on leather of footwear. We carried out a questionnaire survey and grading on the vision and visual-tactile sense perception with 54 pieces of shoe leather samples provided by shoemaking manufacturers. Secondly, a program was written to capture the essential color and adhesion degree of the photo colors of these leather samples as color features. Then, the gray-scale values of the image were analyzed, and the related computational methods of LBP, SCOV, VAR and SAC were put forward on the basis of pixel eight-neighborhood to capture the textural features of the images. It took the captured features of color and texture as input layer and the quantized values of the perceptual words as output layer. Using 49 pieces of leather samples carried out the back propagation neural network training and tested 5 pieces of leather samples. 3 categories with 17 kinds of artificial neural network training verification on the combination forms have been completed. This paper finally found out that the optimal result in the back propagation neural network was both color and VAR texture feature are used as input layer while the quantized values of visual-tactile sense perception are taken as output layer.
The specific achievements of the study were as follows: (1) A parametric transfering method of color and texture features of the shoe leather was proposed. (2) A better Kansei engineering and artificial neural network training method of shoe leather was proposed. An aided design flow of shoe leather with perceptual words and back propagation artificial neural network worked out. (3) A computer-aided evaluation and query system of shoe leather was designed to help the designers to find out the leather sample that is closest to the suggested leather.

中文摘要 I
誌謝 III
目錄 IV
表目錄 VIII
圖目錄 X
第一章 緒論 1
1-1研究背景 1
1-2研究動機 2
1-3研究目的 3
1-4研究範圍與限制 4
1-5研究架構 5
第二章 文獻探討 7
2-1感性工學 7
2-1-1感性工學介紹 7
2-1-2感性工學分類 8
2-2皮革相關研究 9
2-2-1皮革產品種類簡介 9
2-2-2皮革生產流程 10
2-3製鞋業相關研究 11
2-3-1製鞋業概況 11
2-3-2製鞋材料業概況 12
2-4類神經網路 12
2-4-1監督式學習網路 15
2-4-2非監督式學習網路 15
2-5視覺、觸覺相關研究 17
2-5-1視覺 17
2-5-2觸覺 18
2-5-3視觸覺 18
2-6色彩與紋理特徵 19
2-6-1色彩特徵 22
2-6-2紋理特徵 25
第三章 研究方法與步驟 28
3-1倒傳遞類神經網路 28
3-1-1倒傳遞類神經網路訓練 32
3-2語意差異法 33
3-3集群分析 34
3-3-1 K-Means 34
3-4焦點團體法 36
第四章 實驗研究分析與探討 37
4-1皮革樣片收集 39
4-2皮革樣本相關感性語彙蒐集 41
4-2-1 感性語彙蒐集 41
4-2-2感性語彙對挑選 42
4-3感性工學問卷 44
4-3-1感性語彙對意象感覺評分 44
4-4皮革樣片拍攝 45
4-5皮革樣片特徵擷取 45
4-5-1皮革樣片色彩特徵擷取 46
4-5-2皮革樣片紋理特徵擷取 47
4-6集群分析 49
4-6-1集群分析結果 49
4-7倒傳遞類神經網路訓練分類 50
4-8倒傳遞類神經網路建構及訓練 51
4-8-1以色彩特徵為輸入層的視覺BPN-A訓練 53
4-8-2以紋理特徵為輸入層的視覺BPN-B1~B4訓練 56
4-8-3以紋理特徵為輸入層的視觸覺BPN-B5~B8訓練 65
4-8-4以色彩特徵與紋理特徵為輸入層的視覺BPN-C1~C4訓練 74
4-8-5以色彩特徵與紋理特徵為輸入層的視觸覺BPN-C5~C8訓練 83
4-9總結與討論 92
4-9-1 色彩特徵與視覺感性量化值 93
4-9-2 紋理特徵與視覺、視觸覺感性量化值 93
4-9-3 色彩特徵、紋理特徵與視覺、視觸覺感性量化值 94
4-9-4感性語彙對值倒傳遞類神經網路驗證結果 95
4-9-5倒傳遞類神經網路驗證結果誤差率討論 96
第五章 結論與建議 97
5-1研究結果討論 97
5-2研究應用 98
5-3研究貢獻 101
5-4研究後續建議 102
參考文獻 104
【附錄一】 視覺與視觸覺相關語彙每組28對 109
【附錄二】 視覺與視觸覺相關語彙14對 110
【附錄三】 視覺問卷10組感性語彙-受測者印象評分 111
【附錄四】 視觸覺問卷10組感性語彙-受測者感性評分 113
【附錄五】 54組皮革樣片之色彩特徵向量資料 115
【附錄六】 54組皮革樣片之LBP特徵向量資料 116
【附錄七】 54組皮革樣片之SCOV向量資料 117
【附錄八】 54組皮革樣片之VAR向量資料 118
【附錄九】 54組皮革樣片之SAC向量資料 119
【附錄十】 54組皮革樣片色彩與LBP紋理特徵集群分類 120
【附錄十一】 54組皮革樣片色彩與SCOV紋理特徵集群 121
【附錄十二】 54組皮革樣片色彩與VAR紋理特徵集群 122
【附錄十三】 54組皮革樣片色彩與SAC紋理特徵集群 123

1.CCIR. (1990). “Encoding paramenters of digital television for studios” CCIR (Centre for Communication Interface Research) Recommendation 601-2, Int. Radio Consult. Committee, Geneva, Switzerland.
2.Chen, X. J., Shao, F., Barnes, C., Childs, T., & Henson, B. (2009). Exploring Relationships between Touch Perception and Surface Physical Properties. International Journal of Design, 3(2), 67-76.
3.Han, J., Kamber, M., & Pei, J. (2011). Data Mining: Concepts and Techniques (3rd ed.). Morgan Kaufmann Publishers Inc.
4.Hsiao, S. W., Chiu, F. Y., & Lu, S. H. (2010), Product-form design model based on genetic algorithms, International Journal of Industrial Ergonomics 40(3), 237-246.
5.Huang, C. Y. (2016). Face Spoofing Detection from a Single Image Using Texture and Direction Analysis in HSV Color Space. Master dissertation, Department of Electrical and Control Engineering, National Chiao Tung University, 1-55
6.Lai, H. H., Lin, Y. C., Yeh, C. H., & Wei, C. H. (2006). User-oriented design for the optimal combination on product design. International Journal of Production Economics, 100(2), 253-267.
7.Ludden, G.D.S., Schifferstein, H.N.J., &Hekkert P. (2009). Visual-tactual incongruities in products as sources of surprise. Empirical Studies of the Arts ,27(1), 61-87.
8.Ma, M. Y., Chen, C. Y., & Wu, F. G. (2007). A design decision-making support model for customized product color combination. Computers in Industry, 58(6), 504-518.
9.Nagamachi, M. (1995). Kansei engineering: a new ergonomic consumer-oriented technology for product development. International Journal of Industrial Ergonomics, 15(1), 3-11.
10.Ojala, T., Pietikainen, M., & Maenpaa, T. (2002). Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(7), 971-987.
11.Osgood, C. E., & Tzeng, O. (1990). Language, meaning, and culture: The selected papers of CE Osgood. Praeger Publishers.
12.Park, J. H., & Won, S. (2000). Stability analysis for neutral delay-differential systems. Journal of the Franklin Institute, 337(1), 1-9.
13.Pass, G., Zabih, R., & Miller, J. (1997). Comparing images using color coherence vectors. In Proceedings of the fourth ACM International Conference on Multimedia (pp. 65-73). ACM.
14.Rupérez, M. J., Monserrat, C., Alemany, S., Juan, M. C., & Alcañíz, M. (2010). Contact model, fit process and, foot animation for the virtual simulator of the footwear comfort. Computer-Aided Design, 42(5), 425-431.
15.Schifferstein, H. N. J. & Desmet, P. M. A. (2007). The effects of sensory impairments on product experience and personal well-being, Ergonomics, 50(12), 2026-2048.
16.Schultz, L. M. & Petersik, J. T. (1994). Visual-haptic relations in a two-dimensional size- matching task. Perceptual and Motor Skills, 78(2), 395-402.
17.Shieh, M. D., & Yeh, Y. E. (2013). Developing a design support system for the exterior form of running shoes using partial least squares and neural networks. Computers & Industrial Engineering, 65(4), 704-718.
18.Smith, J. R., & Chang, S. F. (1995). Single color extraction and image query. In Image processing, 1995. Proceedings., International conference on Image Processing (3, 528-531). IEEE.
19.Stadtlander, L. & Murdoch, L. (2000). Frequency of occurrence and rankings for touch-related adjectives. Behavior Research Methods, Instruments, and Computers, 32(4), 579-587.
20.Stewart, D. W., Shamdasani, P. N., & Rook, D. W. (1990). Focus groups: Theory and practice. Applied social research methods series. Focus groups: Theory and Practice Applied Social Research Methods Series.
21.Wang, C. C., Yang, C. H., Wang, C. S., Chang, T. R., & Yang, K. J. (2016). Feature recognition and shape design in sneakers. Computers & Industrial Engineering, 102, 408-422.
22.李宏傑（2001）。基於紋理與色彩特徵的影像檢索系統之研究。淡江大學電機工程學系電子電路組碩士論文。
23.長町三生（1999）。淺談感性工學(感性工学のおはなし)。日本東京：日本規格協會。
24.林崇宏（2009)。造形原理：造形與構成設計案例解析。台北全華圖書出版
25.。
26.陳天祥（1998）。感覺與知覺：心理學新論。台北：揚智文化圖書出版。
27.陳靜如（2010）。類神經網路於鞋墊舒適度預測與評價。東海大學工業設計系碩士論文。
28.黃泊鑫（2015）。糖尿病患者之客製化鞋楦設計研究。東海大學工業設計系碩士論文。
29.陳福興（1998）。行職業資訊研發成果專輯。行政院勞工委員會職業訓練局發行。製作中華民國科學學會。
30.黃祥瑋（2010）。二維品質模式與品質機能展開的應用以協助製鞋業產品開發之研究。東海大學工業設計學系碩士論文。
31.黃培恩（2012）。以作業基礎效益評估法建構皮革產業最適皮料投產配置模式。逢甲大學工業工程與系統管理學研究所碩士論文。
32.楊凱傑（2013）。應用類神經網路於球鞋造形曲線認知與設計之研究。東海大學工業設計系碩士論文。
33.呂貞儀（2013）。皮革魅力之研究。東海大學工業設計系碩士論文。
34.鄭瑞傑（2011）。合成皮革在視覺與視觸共覺上之感性研究。成功大學工業設計學系學位論文。
35.呂明泉（2002）。觸覺與視覺對意象差異研究-以塑膠材質咬花為例。成功大學工業設計學系學位論文。
36.柯超茗（1997）。材料視覺與觸覺質感意象的研究。雲林科技大學工業設計系碩士論文。
37.楊東翰（2016）。基於色彩與紋理特徵的超像素追蹤法。國立東華大學資訊工程學系碩士論文。
38.吳俊龍（2004）。以影像紋理與色彩資訊輔助地物分類之研究-以台北地區為例。國立台北大學地政學系碩士論文