arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1709.07337cs.CVstat.ML
收藏

用于细胞检测和分割的高效列生成

Efficient Column Generation for Cell Detection and Segmentation

Universitat Pompeu Fabra(庞培法布拉大学);A&E Technologies(A&E 科技公司);ICREA(加泰罗尼亚高等研究院);Experian Data Lab(益博睿数据实验室)

Chong Zhang, Shaofei Wang, Miguel A. Gonzalez-Ballester, Julian Yarkony

摘要:

研究了细胞密集、紧凑的生物图像中的实例分割问题。我们制定这个任务作为一个整数规划,其中变量对应于细胞和约束强制细胞不重叠。为了解决这个整数规划,我们提出了一个列生成公式,定价程序通过精确优化非常小规模的整数规划来解决。列生成收紧使用奇集不等式,适合优雅的定价问题的优化。我们的列生成方法实现了我们的实例分割问题的快速稳定的任何时候推理。我们在三个不同的光学显微镜数据集上证明,每个数据集有几百个细胞,我们提出的算法迅速达到或超过了最先进的精度。

英文摘要:

We study the problem of instance segmentation in biological images with crowded and compact cells. We formulate this task as an integer program where variables correspond to cells and constraints enforce that cells do not overlap. To solve this integer program, we propose a column generation formulation where the pricing program is solved via exact optimization of very small scale integer programs. Column generation is tightened using odd set inequalities which fit elegantly into pricing problem optimization. Our column generation approach achieves fast stable anytime inference for our instance segmentation problems. We demonstrate on three distinct light microscopy datasets, with several hundred cells each, that our proposed algorithm rapidly achieves or exceeds state of the art accuracy.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑