For a set $S$ of $n$ disjoint line segments in $\mathbb{R}^{2}$, the visibility counting problem is to preprocess $S$ such that the number of visible segments in $S$ from any query point $p$ can be computed quickly. There have been approximation algorithms for this problem with trade off between space and query time. We propose a new randomized algorithm to compute the exact answer of the problem. For any $0<\alpha<1$, the space, preprocessing time and query time are $O_{\epsilon}(n^{4-4\alpha})$, $O_{\epsilon}(n^{4-2\alpha})$ and $O_{\epsilon}(n^{2\alpha})$, respectively. Where $O_{\epsilon}(f(n)) = O(f(n)n^{\epsilon})$ and $\epsilon>0$ is an arbitrary constant number.
翻译:对于以$mathbb{R ⁇ 2}为单位的固定离线线段,可见度计问题在于预处理$S美元,这样就可以从任何查询点迅速计算出以美元计的可见部分数量。对于这个问题,在空间和查询时间之间交换了近似算法。我们建议采用新的随机算法来计算问题的准确答案。对于任何1美元,空间、预处理时间和查询时间为$Oepsilon}(n ⁇ 4-4\alpha})$,(n ⁇ 4-2\alpha})$(n ⁇ 4-2\alpha})美元和$Oepsilon}(n ⁇ 2\alpha})美元。$Oepsilon}(f(n)) =O(f)(n) {epsilon} 美元和 $\psilon>0美元是一个任意不变的数字。