计算机科学与探索
計算機科學與探索
계산궤과학여탐색
JOURNAL OF FRONTIERS OF COMPUTER SCIENCE & TECHNOLOGY
2015年
8期
995-1003
,共9页
贺建峰%符增%易三莉%相艳%崔锐
賀建峰%符增%易三莉%相豔%崔銳
하건봉%부증%역삼리%상염%최예
遗传算法%灰度-梯度熵%多阈值%图像分割%积分图
遺傳算法%灰度-梯度熵%多閾值%圖像分割%積分圖
유전산법%회도-제도적%다역치%도상분할%적분도
genetic algorithm%gray-level&gradient-magnitude entropy%multi-threshold%image segmentation%inte-gral figure
一些基于熵的阈值图像分割技术考虑了空间信息,从而能够提高阈值分割的性能,但是仍然不能较好地区分边缘和噪声。尽管灰度-梯度(gray-level & gradient-magnitude,GLGM)熵算法能有效地解决以上问题,但是针对多目标和复杂图像却不能有效地分割。为此,提出了一种基于遗传算法(genetic algorithm,GA)的GLGM熵多阈值快速分割方法。该方法应用积分图思想将GLGM熵算法阈值搜索空间从O(9′ L)降到O(L),并将GLGM熵算法从单阈值拓展到多阈值。最后应用基于实数编码的遗传算法搜索GLGM熵多阈值的最佳阈值。仿真结果表明,该方法能够实现图像的快速多阈值分割,适合复杂图像分割。
一些基于熵的閾值圖像分割技術攷慮瞭空間信息,從而能夠提高閾值分割的性能,但是仍然不能較好地區分邊緣和譟聲。儘管灰度-梯度(gray-level & gradient-magnitude,GLGM)熵算法能有效地解決以上問題,但是針對多目標和複雜圖像卻不能有效地分割。為此,提齣瞭一種基于遺傳算法(genetic algorithm,GA)的GLGM熵多閾值快速分割方法。該方法應用積分圖思想將GLGM熵算法閾值搜索空間從O(9′ L)降到O(L),併將GLGM熵算法從單閾值拓展到多閾值。最後應用基于實數編碼的遺傳算法搜索GLGM熵多閾值的最佳閾值。倣真結果錶明,該方法能夠實現圖像的快速多閾值分割,適閤複雜圖像分割。
일사기우적적역치도상분할기술고필료공간신식,종이능구제고역치분할적성능,단시잉연불능교호지구분변연화조성。진관회도-제도(gray-level & gradient-magnitude,GLGM)적산법능유효지해결이상문제,단시침대다목표화복잡도상각불능유효지분할。위차,제출료일충기우유전산법(genetic algorithm,GA)적GLGM적다역치쾌속분할방법。해방법응용적분도사상장GLGM적산법역치수색공간종O(9′ L)강도O(L),병장GLGM적산법종단역치탁전도다역치。최후응용기우실수편마적유전산법수색GLGM적다역치적최가역치。방진결과표명,해방법능구실현도상적쾌속다역치분할,괄합복잡도상분할。
Due to considering the gray level spatial distribution information, some image segmentation technologies based on entropy threshold can enhance the thresholding segmentation performance. However, they still cannot dis-tinguish image edges and noise well. Even though GLGM (gray-level&gradient-magnitude) entropy can effectively solve the problem, it cannot segment effectively multi-objective and complex image. So, this paper proposes image segmentation with multi-threshold of GLGM entropy based on genetic algorithm. In the proposed method, integral figure is introduced in order to make threshold searching dimension from original O(9 ′ L) to O(L) , and the single threshold segmentation of GLGM entropy is further extended to multi-threshold segmentation. Lastly, the real-code-GA is used to search the best thresholds. The simulation results show that this method can be effectively applied for the multi-threshold segmentation of complex images.