研究生: |
陳柏翰 Bo-Han Chen |
---|---|
論文名稱: |
植基於支持向量機之快速彩色濾波陣列樣型辨識方法 Fast SVM-based identification of arbitrary CFA images |
指導教授: |
鍾國亮
Kuo-Liang Chung |
口試委員: |
廖弘源
Mark Liao 范國清 Kuo-Chin Fan 貝蘇章 Soo-Chang Pei 徐繼聖 Gee-Sern Hsu |
學位類別: |
碩士 Master |
系所名稱: |
電資學院 - 資訊工程系 Department of Computer Science and Information Engineering |
論文出版年: | 2016 |
畢業學年度: | 104 |
語文別: | 中文 |
論文頁數: | 30 |
中文關鍵詞: | 機器學習 、支持向量機 、交叉驗證 、馬賽克影像 、彩色濾波陣列 |
外文關鍵詞: | Machine learning, Support vector machine, Cross-validation, Mosaic image, Color Filter Array |
相關次數: | 點閱:191 下載:2 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
馬賽克影像中,每一個通過彩色濾波陣列的像素點僅由一個顏色所組成,遺
失彩色濾波陣列將無法解馬賽克。本篇論文提出一套植基於支持向量機的彩色濾
波陣列影像辨識方法。本篇論文所提出方法能辨識十一種彩色濾波陣列,方法分
為兩階段。第一階段藉由在空間域抽取彩色濾波陣列特徵訓練與預測支持向量機,
能直接辨識六種彩色濾波陣列,其餘的五種因為有誤判發生,需要在第二階段進
行辨識。第二階段使用決定樹能夠正確辨識其餘五種彩色濾波陣列。在本實驗中,
我們實五組作交叉驗證,實驗結果顯示和先前方法相比,本篇論文所提植基於支
持向量機之二階段辨識方法與先前方法相比,能辨識在更短的時間內辨識彩色濾
波陣列種類。
Considering mosaic images, each pixel captured by color filter array is composed
of only one primary color. Without the color filter array (CFA) pattern information, it
is hard to demosaic or compress CFA images. In this paper, we propose a SVM based
method to identify the CFA pattern of the input CFA image. The proposed method has
two stages. In the first stage, we train SVM by features extracted from the spatial
domain of CFAs. In this stage, 6 CFA structures can be recognized and the other 5 CFA
structures can be identified in the second stage. In the second stage, we use decision
tree approach to identify the remaining CFA structures. Based on 5 groups crossvalidation, experimental results demonstrate that the proposed SVM based
identification method can identify CFA structures faster when compared with the stateof-the-art algorithm.
[1] R. Lukac, Single-Sensor Imaging: Methods and Applications for Digital Cameras,
CRC Press/Taylor and Francis, Boca Raton, FL, 2008
[2] B. E. Bayer, “Color imaging array,” U.S. Patent# 3 971 065, 1976.
[3] M. Parmar, and S. J. Reeves, “A perceptually based design methodology for color
filter arrays,” Proceeding of IEEE ASSP 2004, Montreal, pp. 473-476, May 2004.
[4] R. Lukac, and K. N. Plataniotis, “Color filter arrays: Design and performance
analysis,” IEEE Trans. Consumer Electron, Vol 51, No. 4, pp. 1260-1267, Nov. 2005.
[5] J. E. Adams, and J. F. Hamilton, “Adaptive color plan interpolation in single sensor
color electric camera,” U.S. Patent# 5 506 619, 1996.
[6] W. Pennebaker, J. Mitchell, JPEG, “Still Image Data Compression
Standard,” Van Nostrand, New York, 1993.
[7] Y.H. Huang, K.L. Chung and T.J. Lin, “Efficient identification of arbitrary color
filter array images based on the frequency domain approach,” Signal Processing,
Volume 115, October 2015, Pages 120–129.
[8] [Online]. Available: <https://www.csie.ntu.edu.tw/~cjlin/>, slate March 5 2016.
[9] M. Mignotte, “Segmentation by fusion of histogram-based k-means clusters in
different color spaces,” IEEE Trans. Image Process. 17 (May (5)) (2008) 780–787.
[10][Online].Available:<http://r0k.us/graphics/kodak/>, slate March 5 2016.
[11][Online].Available:<http://www4.comp.polyu.edu.hk/~cslzhang/CDM_Dataset.ht
m>, slate March 5 2016.
[12][Online].Available:<https://tecnick.com/public/code/cp_dpage.php?aiocp_dp=test
images >, slate March 5 2016.
[13][Online].Available:<http://tabby.vision.mcgill.ca/html/browsedownload.html>,
slate March 5 2016.