貝蘇章臺灣大學：電信工程學研究所魏郁珊Wai, Yu-ShanYu-ShanWai2007-11-272018-07-052007-11-272018-07-052006http://ntur.lib.ntu.edu.tw//handle/246246/58815數十年來，自動影像辨識一直是電腦視覺領域中十分渴切希望解決的課題。自動影像辨識，類比於人類對影像的知覺，第一步必須將影像中均質的區塊做分割，才能進一步討論該區塊的形狀質地色彩進而進行辨識。然而，就如同自動影像辨識一般，雖然自動影像分割技術及演算法不斷在進步，卻仍是個尚未完整解決的問題。這是因為一般影像內容的不確定性和複雜程度較高。不過大致上而言，基本影像處理以及樣式辨認技術，將有助於簡化影像切割問題的複雜程度並幫助呈現出更好的切割效果。 論文前半將介紹截至目前為止成效較好的兩種影像切割演算法，實作其演算法並且對架構加以修改，增進效率並且同時保持解析度和準確度。此外，我們也對某些較難切割的影像嘗試了幾種增進準確度的方法，並得到不錯的效果。 若從另外一個角度來看影像切割的問題，既然自動辨識和切割是如此困難，何不退一步，藉由使用者的幫助來解決問題呢? 於是，在論文的後半部份，我們簡短地介紹了目前某些互動式物件擷取系統的概念以及操作方式，並提出一個新的系統架構。使用者只需要做一些簡單的指定，系統便可以快速地找出使用者所想要的物體。這都要感謝使用者的幫助給予了系統足夠的資訊來降低問題的複雜性和不確定性。 在這論文裡所提出的影像切割架構和互動式物件擷取架構都應用到了多層比例 (Multi-Scale) 架構的概念。The automatic recognition of images has been a researched topic over decades yet still a difficult task to accomplish. Similar as the process of human perception, the first step of automatic recognition should be image segmentation to segment different homogeneous patches into regions. However, like the automatic recognition, segmentation remains a yet satisfactorily solved problem. This is due to the uncertainty and complexity nature of images. By the aid of various image processing techniques and pattern recognition analysis, this problem may reduced to some less complicated level, thus helps to achieve preferable result. We first introduce two excellently performed algorithms of color image segmentation up to date, and made some modification on the structural phase and improved their efficiency while preserving resolution and accuracy. Moreover, we experimented on adding some special approaches as pre-processing, tested on specific images which are difficult to segment and get very good results. From another aspect to look upon this problem, since the recognition and also the segmentation so difficult a problem to solve, why not ask for a little assistance of human specification? In the latter part of the thesis, interactive object extraction is briefly covered and a new system we proposed is tested. With simple user specification, we could solve the extraction problem by a simpler and faster version of solution. This reduction of complexity should thanks to the prior knowledge given by human specification. Both the automatic image segmentation and interactive object extraction take the advantage of multi-scale structure.CHAPTER 1 Introduction………………………………………………………………….. 1 CHAPTER 2 Basic Properties and Processing of Image………………………………….. 3 2.1 Color Information and Processing………………………………………………………. 3 2.1.1 Popular Color Spaces………………………………………………………………………. 4 2.1.1.1 CIE Tristimulus Color Coordinates………………………………………………….. 4 2.1.1.2 Device dependent RGB & CMYK color space……………………………………... 7 2.1.2 Useful Techniques on Color Information Processing…………………………………… 8 2.1.2.1 Color Histogram……………………………………………………………………….. 8 2.1.2.2 Filtering…………………………………………………………………………………. 9 2.2 Texture…………………………………………………………………………………… 10 2.3 Shape…………………………………………………………………………………….. 10 2.4 Conclusion………………………………………………………………………………. 11 CHAPTER 3 Color Image Segmentation Algorithms with Implementation and Experiments………………………………………………………………….. 12 3.1 Concepts of Current Popular Methods…………………………………………………... 12 3.1.1 Morphological Watersheds…………………………………………………………………. 12 3.1.2 Region Based Segmentation……………………………………………………………….. 15 3.1.2.1 Data Clustering………………………………………………………………………… 15 3.1.2.2 Region Merging………………………………………………………………………… 21 3.1.3 Other Segmentation Techniques…………………………………………………………… 26 3.2 Useful techniques for Improvements on Image Segmentation………………………….. 28 3.2.1 Texture Smoothing…………………………………………………………………………… 28 3.2.2 Highlights, Shadow, Shading, and Color Constancy…………………………………… 30 3.3 Implementation of Popular Image Segmentation Algorithms with Modification and Improvements………………………………………………..…………………….……… 30 3.3.1 Implementation of Mean-Shift Image Segmentation Algorithm……………………….. 30 3.3.2 Implementation of Statistical Region Merging Algorithm……………………………… 41 3.4 Applications……………………………………………………………………………... 45 3.5 Conclusion………………………………………………………………………………. 46 CHAPTER 4 Interactive Object Extraction……………………………………………….. 49 4.1 Introductory Overview on Popular Related works……………………………………… 50 4.2 Proposed System………………………………………………………………………… 54 4.3 Conclusion………………………………………………………………………………. 62 CHAPTER 5 Conclusion and Future Work……………………………………………….. 63 5.1 Conclusion………………………………………………………………………………. 63 5.2 Future Work……………………………………………………………………………… 64 REFERENCE………………………………………………………………………………... 667692450 bytesapplication/pdfen-US影像切割色彩影像處理物件擷取image segmentationimage processinginteractive object extractionthesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/58815/1/ntu-95-R93942049-1.pdf