Identification of Green, Oolong and Black Teas in China via Wavelet Packet Entropy and Fuzzy Support Vector Machine
To develop an automatic tea-category identification system with a high recall rate, we proposed a computer-vision and machine-learning based system, which did not require expensive signal acquiring devices and time-consuming procedures. We captured 300 tea images using a 3-CCD digital camera, and th...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2015-09-01
|
| Series: | Entropy |
| Subjects: | |
| Online Access: | http://www.mdpi.com/1099-4300/17/10/6663 |
| _version_ | 1798042581304606720 |
|---|---|
| author | Shuihua Wang Xiaojun Yang Yudong Zhang Preetha Phillips Jianfei Yang Ti-Fei Yuan |
| author_facet | Shuihua Wang Xiaojun Yang Yudong Zhang Preetha Phillips Jianfei Yang Ti-Fei Yuan |
| author_sort | Shuihua Wang |
| collection | DOAJ |
| description | To develop an automatic tea-category identification system with a high recall rate, we proposed a computer-vision and machine-learning based system, which did not require expensive signal acquiring devices and time-consuming procedures. We captured 300 tea images using a 3-CCD digital camera, and then extracted 64 color histogram features and 16 wavelet packet entropy (WPE) features to obtain color information and texture information, respectively. Principal component analysis was used to reduce features, which were fed into a fuzzy support vector machine (FSVM). Winner-take-all (WTA) was introduced to help the classifier deal with this 3-class problem. The 10 × 10-fold stratified cross-validation results show that the proposed FSVM + WTA method yields an overall recall rate of 97.77%, higher than 5 existing methods. In addition, the number of reduced features is only five, less than or equal to existing methods. The proposed method is effective for tea identification. |
| first_indexed | 2024-04-11T22:37:32Z |
| format | Article |
| id | doaj.art-223852daa666461aa157e1e866926fa5 |
| institution | Directory Open Access Journal |
| issn | 1099-4300 |
| language | English |
| last_indexed | 2024-04-11T22:37:32Z |
| publishDate | 2015-09-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Entropy |
| spelling | doaj.art-223852daa666461aa157e1e866926fa52022-12-22T03:59:09ZengMDPI AGEntropy1099-43002015-09-0117106663668210.3390/e17106663e17106663Identification of Green, Oolong and Black Teas in China via Wavelet Packet Entropy and Fuzzy Support Vector MachineShuihua Wang0Xiaojun Yang1Yudong Zhang2Preetha Phillips3Jianfei Yang4Ti-Fei Yuan5School of Computer Science and Technology, Nanjing Normal University, 210023 Nanjing, ChinaDepartment of Mathematics and Mechanics, China University of Mining and Technology, 221008 Xuzhou, ChinaSchool of Computer Science and Technology, Nanjing Normal University, 210023 Nanjing, ChinaSchool of Natural Sciences and Mathematics, Shepherd University, Shepherdstown, 25443 West Virginia, WV, USAJiangsu Key Laboratory of 3D Printing Equipment and Manufacturing, 210042 Nanjing, ChinaSchool of Psychology, Nanjing Normal University, 210008 Nanjing, ChinaTo develop an automatic tea-category identification system with a high recall rate, we proposed a computer-vision and machine-learning based system, which did not require expensive signal acquiring devices and time-consuming procedures. We captured 300 tea images using a 3-CCD digital camera, and then extracted 64 color histogram features and 16 wavelet packet entropy (WPE) features to obtain color information and texture information, respectively. Principal component analysis was used to reduce features, which were fed into a fuzzy support vector machine (FSVM). Winner-take-all (WTA) was introduced to help the classifier deal with this 3-class problem. The 10 × 10-fold stratified cross-validation results show that the proposed FSVM + WTA method yields an overall recall rate of 97.77%, higher than 5 existing methods. In addition, the number of reduced features is only five, less than or equal to existing methods. The proposed method is effective for tea identification.http://www.mdpi.com/1099-4300/17/10/6663tea identificationwavelet packet entropyShannon entropywavelet analysissupport vector machine (SVM)fuzzy SVMinformation theory |
| spellingShingle | Shuihua Wang Xiaojun Yang Yudong Zhang Preetha Phillips Jianfei Yang Ti-Fei Yuan Identification of Green, Oolong and Black Teas in China via Wavelet Packet Entropy and Fuzzy Support Vector Machine Entropy tea identification wavelet packet entropy Shannon entropy wavelet analysis support vector machine (SVM) fuzzy SVM information theory |
| title | Identification of Green, Oolong and Black Teas in China via Wavelet Packet Entropy and Fuzzy Support Vector Machine |
| title_full | Identification of Green, Oolong and Black Teas in China via Wavelet Packet Entropy and Fuzzy Support Vector Machine |
| title_fullStr | Identification of Green, Oolong and Black Teas in China via Wavelet Packet Entropy and Fuzzy Support Vector Machine |
| title_full_unstemmed | Identification of Green, Oolong and Black Teas in China via Wavelet Packet Entropy and Fuzzy Support Vector Machine |
| title_short | Identification of Green, Oolong and Black Teas in China via Wavelet Packet Entropy and Fuzzy Support Vector Machine |
| title_sort | identification of green oolong and black teas in china via wavelet packet entropy and fuzzy support vector machine |
| topic | tea identification wavelet packet entropy Shannon entropy wavelet analysis support vector machine (SVM) fuzzy SVM information theory |
| url | http://www.mdpi.com/1099-4300/17/10/6663 |
| work_keys_str_mv | AT shuihuawang identificationofgreenoolongandblackteasinchinaviawaveletpacketentropyandfuzzysupportvectormachine AT xiaojunyang identificationofgreenoolongandblackteasinchinaviawaveletpacketentropyandfuzzysupportvectormachine AT yudongzhang identificationofgreenoolongandblackteasinchinaviawaveletpacketentropyandfuzzysupportvectormachine AT preethaphillips identificationofgreenoolongandblackteasinchinaviawaveletpacketentropyandfuzzysupportvectormachine AT jianfeiyang identificationofgreenoolongandblackteasinchinaviawaveletpacketentropyandfuzzysupportvectormachine AT tifeiyuan identificationofgreenoolongandblackteasinchinaviawaveletpacketentropyandfuzzysupportvectormachine |