選擇性背景優(yōu)先的顯著性檢測模型
doi: 10.11999/JEIT140119 cstr: 32379.14.JEIT140119
基金項(xiàng)目:
國家自然科學(xué)基金(90920013)資助課題
Saliency Detected Model Based on Selective Edges Prior
-
摘要: 在檢測圖像顯著性區(qū)域的領(lǐng)域中，背景優(yōu)先是一個(gè)較新的思路，但會遇到背景鑒別這個(gè)具有挑戰(zhàn)性的難題。該文提出背景真實(shí)性的判斷問題，在探索的過程中發(fā)現(xiàn)背景通常具有連續(xù)性的特征，根據(jù)這一特性實(shí)現(xiàn)了判定背景的方法，并將判斷的結(jié)果作為顯著性先驗(yàn)值應(yīng)用于后繼的計(jì)算中，最終結(jié)果的準(zhǔn)確性和正確性得到有效提高。該文首先采用均值漂移(MS)分割算法將圖片預(yù)分為超像素，計(jì)算所有超像素的初始顯著值；隨后提取原圖的4個(gè)邊界條，計(jì)算每兩條之間的色彩直方圖距離，判定小于預(yù)設(shè)閾值的兩條邊界作為真的背景，選擇它們作為優(yōu)先邊界，計(jì)算先驗(yàn)顯著性值；最后進(jìn)行顯著性計(jì)算，得到最終的顯著圖。實(shí)驗(yàn)結(jié)果表明，該算法能夠準(zhǔn)確檢測出顯著性區(qū)域，與其他6種算法相比具有較大優(yōu)勢。
-
關(guān)鍵詞:
- 計(jì)算機(jī)視覺 /
- 顯著性分析 /
- 背景連續(xù)性 /
- 色彩直方圖 /
- 超像素
Abstract: In the field of saliency detection, background prior has become a novel viewpoint, but how to identify the real background is challenging. In this paper, a background-identified method is proposed based on homology continuity using the extracted background features, and the identified background is applied to the following computation, improving the eventual saliency map in accuracy as well as correctness. First, the primary saliency of each superpixel produced by Mean Shift (MS) segmentation algorithm is calculated. Second, 4 edges are extracted to generate their RGB histograms, and the Euclidean distance between each two of the histograms is calculated, if the distance is smaller than a given value, these two edges are defined to be continual and more likely to be the real background. Finally, the pixels saliency is calculated using the prior background knowledge to figure the final saliency map. The results show that the proposed method outperforms other algorithms in accuracy and efficiency.-
Key words:
- Computer vision /
- Saliency analysis /
- Background continuity /
- RGB histogram /
- Super pixel
-
計(jì)量
- 文章訪問數(shù): 2097
- HTML全文瀏覽量: 187
- PDF下載量: 1152
- 被引次數(shù): 0
下載:
