首页> 外文会议>International conference on swarm intelligence >Improved Two-Dimensional Otsu Based on Firefly Optimization for Low Signal-to-Noise Ratio Images
【24h】

Improved Two-Dimensional Otsu Based on Firefly Optimization for Low Signal-to-Noise Ratio Images

机译:基于萤火虫优化的改进二维Otsu用于低信噪比图像

获取原文

摘要

To improve two-dimensional (2D) Otsu thresholding's performance in both computation speed and segmentation quality, an improved 2D Otsu algorithm is proposed for low Signal-to-noise Ratio (SNR) images. A new 2D histogram is defined based on median gray-scale and Gaussian average gray-scale. By meeting better to the assumption of that the object's probability and the background's probability sum up to 1, the new 2D histogram enhances the thresholding algorithm's robustness to severe noise. Then a scheme of calculating the fitness function based on firefly optimization algorithm is employed to search for optimal thresholds. The proposed algorithm is applied to typical low SNR images-microscopic images of ocean plankton, and to Lenna test image. Experiment results show that with better thresholding quality, the running time of the proposed algorithm is reduced to 2.5% of the conventional 2D Otsu.
机译:为了提高二维(2D)Otsu阈值处理在计算速度和分割质量上的性能,针对低信噪比(SNR)图像提出了一种改进的2D Otsu算法。基于中值灰度和高斯平均灰度定义了新的2D直方图。通过更好地满足物体的概率和背景的概率之和为1的假设,新的2D直方图增强了阈值算法对严重噪声的鲁棒性。然后采用基于萤火虫优化算法的适应度函数计算方案来寻找最优阈值。该算法适用于海洋浮游生物的典型低信噪比图像-显微图像,以及莱纳测试图像。实验结果表明,以更好的阈值质量,该算法的运行时间减少到传统2D Otsu的2.5%。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号