室內與室外影像分類之研究

Indoor vs Outdoor Image Classification of Researches

柯品名(Ping-ming Ke)

國立虎尾科技大學/電機資訊學院/光電與材料科技研究所/碩士(2005年)

摘要

近年來，隨著網際網路的盛行、各種資料儲存容量的增大及數位相機的普及化，所以數位影像資料量也日益龐大。因此，以內容為基礎之圖片擷取與分類逐漸成為近年來許多研究的主題。
以往貝氏分類器應用在室內與室外圖片分類的問題上，往往只能夠依據室外圖片中藍色的天空來分類出室外照片，對於室外照片是花草樹木等景物的圖片，就沒有辦法準確的分類。因此本篇論文，便保留了利用YUV分類有藍色天空的室外照，也利用EDH entropy以及ECV entropy這二個特徵來找出有花草樹木的室外圖片，藉此提升整個系統的分類正確率。
在本篇論文中，除了利用YUV、EDH entropy和 ECV entropy之外，另外也加入了HSI這個特徵，來偵測出更多的圖片，最後再利用結合色彩分佈及邊緣方向樹狀結構分類器，對所有的特徵作特徵選取的動作，藉此選出最有效的特徵，在經過實驗後，本論文所提出之方法與其他方法比較，確實有較佳的分類準確度。

並列摘要

In the early, with the growth of the World Wide Web, the improved storage techniques and popularity of digital images have led to the proliferation of images. Contented-based image retrieval and classification have become important research issues in the last few years.
The Bayes Classification was used on the issue that Indoor vs Outdoor Image Classification, can only classify the outdoor image according to the blue sky in the outdoor image often, to the thing that the photo is the images of scenery , such as flowers plants and trees ,etc. There is accurate classification on outdoor image. In our research , keep and utilize YUV to classify the outdoors of the blue sky, make use of ECV entropy, this characteristic, to find out the outdoor image with flowers plants and trees , raise the hit rate of classification of the whole system.
In this paper, besides utilizing YUV、EDH entropy and ECV entropy, Combination this feature of HSI in addition, to detect and measure more image , utilize and combine color distributing and tree structure Classification of edge direction , do movements that choose of the feature to all features, elect the most effective feature by this, raise the hit rate of the whole system. Through experimental evaluations, our method is shown to have higher accuracy in classifying images than other tested methods.

參考文獻

[1] S. F. Chang, W. Chen, H. J. Meng, H. Sundaram, and D. Zhong, “A Fully Automated Content-Based Video Search Engine Supporting Spatiotemporal Queries,” IEEE Transactions on Circuit and Systems for Video Technology, Vol. 8, No. 5, September 1998.

[5] Yue Zhang, Mario A. Nascimento, and Osmar R. Zaïane, “Building Image Mosaics:An Application of Content-based Image Retrieval,” IEEE International Conference on Multimedia and Expo, Baltimore, MD, USA, July 6-9, 2003.

[6] J.Li, A. Najmi, and R.M.Gray, “Image classification by a two-dimension hidden Markov model,” IEEE Transaction on Signal Processing, vol. 48, no. 2, pp. 517-533, February 2000.

[7] M. Szummer and R.W. Picard, “Indoor-Outdoor Image Classification,” IEEE International Workshop on Content-based Access of Image and Video Databases, in conjunction with ICCV'98. Bombay, India, 1998.

[10] A. Barla, F. Odone, and A. VerriOld, “Fashioned State-of-the-Art Image Classification,” ICIAP 2003.