We introduce a High Dynamic Range (HDR) tone mapping technique that preserves "fine" image features that have significantly different luminances than their surrounding areas, most notably sharp specular highlights and significant textures. This work enhances the family of techniques derived from Durand and Dorsey 02, which utilize bilateral filtering to separate large scale features from detail features. Using the low noise content of HDR imagery to our advantage, we introduce the concept of utilizing shot noise detection (noise outlier rejection) techniques to alternatively classify fine features. This is accomplished through the Rank Order Absolute Difference statistic discussed by Garnett et al 05.