SAR Image Segmentation Algorithm Using Hierarchical Region Merging with Edge Penalty
-
摘要: 該文利用方向邊緣強(qiáng)度信息,建立一種新的邊緣懲罰SAR圖像分割模型,提出一種最小化該模型的層次區(qū)域合并算法。利用多方向比例邊緣檢測(cè)算子(MDRED)提取邊緣強(qiáng)度信息,并結(jié)合分水嶺變換獲得高質(zhì)量的初始過(guò)分割結(jié)果。利用多邊形近似區(qū)域邊緣,提取邊緣的方向,將方向邊緣強(qiáng)度映射(OESM)融入邊緣懲罰中,獲得懲罰強(qiáng)度與邊緣強(qiáng)度呈反比的懲罰項(xiàng)。逐漸增大邊緣懲罰項(xiàng)的強(qiáng)度,獲得由圖像特征驅(qū)動(dòng)的層次區(qū)域合并算法。利用區(qū)域鄰接圖(RAG)表示圖像分割,提高區(qū)域合并的速度。實(shí)驗(yàn)表明:該文方法與其它方法相比在性能和效率上都有優(yōu)勢(shì),獲得更好的分割結(jié)果。
-
關(guān)鍵詞:
- SAR圖像 /
- 圖像分割 /
- 邊緣懲罰 /
- 層次區(qū)域合并 /
- 方向邊緣強(qiáng)度映射
Abstract: A new SAR image segmentation model with edge penalty is constructed, which uses oriented edge strength information and a minimized hierarchical region merging algorithm is proposed in this paper. The edge strength information is extracted by using Multi-Direction Ratio Edge Detector (MDRED), based on which a high quality initial over-segmentation is obtained using watershed transformation. In order to extract the directions of boundaries of regions, polygons are used to approximate them, and a penalty term whose power is in inverse proportion to edge strength is obtained by incorporating Oriented Edge Strength Map (OESM) into the term. A hierarchical region merging algorithm driven by image features is obtained through graduated increased edge penalty. In order to accelerate the region merging, the Region Adjacency Graph (RAG) is used to represent the image segmentation. The experimental results show that the proposed method has advantages in performance and efficiency, and obtains better segmentation results with respect to other methods. -
計(jì)量
- 文章訪問(wèn)數(shù): 1997
- HTML全文瀏覽量: 190
- PDF下載量: 338
- 被引次數(shù): 0