內容簡介
本書提出了一種系統化的女性面部膚色探索＆分類方法，並將其應用於化妝品市場。該方法是對人臉識別理論的一種擴展。在資料獲取過程中，將RGB格式的圖像轉換為Lab空間格式，然後利用模糊C均值理論對資料進行聚類和分組。結果按實驗資料進行分類和分組。創建了兩個程式。第一個程式“FaceRGB”，自動從圖像中捕捉顏色。第二個程式“ColorFCM”，對膚色資訊進行聚類和分組。研究結果可用於幫助專家系統在化妝和新產品開發過程中選擇定制顏色。最後，本書結合默克的色彩趨勢預測，將聚類後的膚色與默克的六種膚色理念相結合進行配對，將結果應用於化妝品中，更清楚地認識到研究的價值和應用的未來發展。
本書特色
消費者行為是複雜的，如何掌握流行趨勢？
在化妝品市場，個人的直覺和色彩選擇的時尚趨勢是消費者的準則。
本書運用默克的色彩趨勢預測，展現AI智慧在時尚趨勢中更精彩的價值。
作者介紹
作者簡介
顏志晃 博士 Chih-Huang Yen, Ph. D.
[學經歷]
台灣成功大學 工業設計研究所 博士
中國閩南師範大學 藝術學院 副教授
台灣高雄師範大學 兼任助理教授
台灣台鉅集團 設計總監
法商伯納集團 格尚公司 開發設計經理
[研究領域]
色彩學、色彩管理、流行趨勢研究分析、模糊理論應用、產品研究與開發、閩南文化創意設計、產品開發流程研究、人工智慧設計應用、大數據的設計分析與應用
Associated professor in the Art school of Minnan Normal University, Zhangzhou, Fujian, China. He is a Ph.D. in the Department of Industrial Design at the National Cheng Kung University, Tainan, Taiwan, Republic of China. He was a design director in Tair Jiuh Enterprise., co. (Taiwan) for cosmetic industry. He spent about 11 years in SAGA company (France) as a R&D manager did watch & jewelry design and development. His major research interests include application of fuzzy set theory on product design, and gray theory to product design, color planning for product design, and design management.
目錄
PREFACE
SUMMARY
ACKNOWLEDGEMENTS
LIST OF SYMBOLS AND ABBREVIATIONS
CHAPTER 1 INTRODUCTION
1.1 Related research
1.2 Market Assessment from Merck
1.3 Outline of this study
CHAPTER 2 LITERATURE REVIEW
2.1 The choice for colour space
2.1.1 RGB colour model
2.1.2 CIELAB colour model
2.1.3 RGB & CIELAB conversions
2.2 Taguchi Methods
2.2.1 Basic concept
2.2.2 Properties of an orthogonal array
2.2.3 Assumptions of the Taguchi method
2.2.4 Designing an experiment
2.3 Fuzzy C-means
2.4 Facial recognition system
CHAPTER 3 IMPLEMENTATION METHOD
3.1 6 points colour detection
3.1.1 Hypothesis _ Captured colour from 6 points
3.1.2 Taguchi method finds optimization
3.2 Verification for 6 points to detect facial colour
3.2.1 Spectrometer Application
3.2.2 Test Design
3.2.3 Test comparison of FaceRGB and FOS (Fiber of spectrometer)
CHAPTER 4 CASE STUDY
4.1 FaceRGB
4.1.1 Outliers
4.1.2 FaceRGB operation
4.2 ColorFCM
4.2.1 Expectation Maximization
4.2.2 E-step (Expectation step)
4.2.3M-step (Maximization)
4.2.4 K-means++
4.2.5 ColorFCM implementation
CHAPTER 5 RESULTS & DISCUSSION
5.1 Experimental verification
5.2 RGB & YCbCr conversions
5.3 Training samples for FaceRGB
5.4 RGB with a large quantity of images
5.5 ColorFCM result by Fuzzy C-means
CHAPTER 6 APPLICATION
6.1 Connection between the colour clusters and Merck makeup trends
6.2 For personal application
6.3 For trend applications
6.4 Conclusion
REFERENCES
