結合紋理資訊於使用超級像素的非監督式影像區域切割

Abstract

Segmentation by Aggregating Superpixels (SAS) [2] 是一個近年新提出的非監督式的影像切割架構，展現了很好的潛力與發展性。然而在SAS的架構中，並沒有使用到已於許多研究 [14-18] 顯現出效益的紋理資訊。因此本論文提出一有效的方法以使用超級像素 (superpixels) 結合紋理資訊於SAS的影像切割架構之中。為了擷取出紋理資訊，演算法首先進行紋理濾波 (texture filtering) 和高斯混合模型分群 (Gaussian Mixture Models (GMM) clustering)，再由發展出的邊緣意識性低通濾波 (edge-aware low-pass filtering) 產生出多種精細等級的紋理超級像素 (texture superpixels)，最後，藉由聯合紋理超級像素與原本於 [2] 使用的超級像素集合，本篇論文可以成功的將紋理資訊與SAS影像切割架構做結合。經由實驗證明，在有公信力的Berkeley Segmentation Dataset (BSDS)、使用三項通用的區域相鄰特性的評分標準 (region-based evaluation criteria)，本篇論文演算法相較於先前研究 [2] 與其他基準影像切割方法 [5, 6, 9] 可以達到更好的表現。

Recently, an unsupervised image segmentation framework, Segmentation by Aggregating Superpixels (SAS) [2] is proposed and shown to be very promising. However, the texture cues, which have been shown to be very effective in many researches [14-18], are absent in [2]. In this thesis, we propose an effective method for incorporating texture information into the SAS framework, using superpixels. To extract texture information, our algorithm first uses texture filtering and subsequent Gaussian Mixture Models (GMM) clustering which is modified from [16]. Then, we develop an edge-aware low-pass filtering to generate multiple-scale texture superpixels from GMM clustering results. Finally, by joining texture superpixels with the superpixel set originally used in [2], the incorporation of texture information is accomplished. Our method achieves superior performance on the well-known Berkeley Segmentation Dataset (BSDS) under multiple prevailing region-based evaluation criteria when compared to other benchmark algorithms.