We present a refined analysis of 15,038 Kepler main sequence light curves to determine the stellar rotation periods. The initial period estimates come from an autocorrelation function, as has been done before. We then measure the duration of every intensity dip in the light curve, expressed as fractions of the initial rotation period estimate. These dip duration distributions are subdivided into several regions whose relation to each other helps determine which harmonic of the initial rotation period is most physically plausible. We compare our final rotation periods to those from McQuillan et al. (2014) and find that the great majority agree, but about 10% of their periods are doubtful (usually twice as long as is most plausible). We are still refining our method, and will later extend it to more stars to substantially increase the sample of reliable stellar rotation periods.

中文翻译：

---

两次浸渍以完善恒星旋转周期

我们对15,038个开普勒主序列光曲线进行了精确分析，以确定恒星旋转周期。像以前一样，初始周期估算值来自自相关函数。然后，我们测量光曲线中每个强度下降的持续时间，以初始旋转周期估算值的分数表示。这些浸入持续时间分布可细分为几个区域，它们之间的关系有助于确定初始旋转周期的哪个谐波在物理上最合理。我们将最终轮换周期与McQuillan等人的轮换周期进行比较。（2014年），发现绝大多数人都同意，但是大约有10％的时期值得怀疑（通常是最合理时期的两倍）。我们仍在完善我们的方法，