Herschel has revolutionized our ability to measure column densities (N_H_) and temperatures (T) of molecular clouds thanks to its far infrared multiwavelength coverage. However, the lack of a well defined background intensity level in the Herschel data limits the accuracy of the NH and T maps. We aim to provide a method that corrects the missing Herschel background intensity levels using the Planck model for foreground Galactic thermal dust emission. For the Herschel/PACS data, both the constant-offset as well as the spatial dependence of the missing background must be addressed. For the Herschel/SPIRE data, the constant-offset correction has already been applied to the archival data so we are primarily concerned with the spatial dependence, which is most important at 250um. We present a Fourier method that combines the publicly available Planck model on large angular scales with the Herschel images on smaller angular scales. We have applied our method to two regions spanning a range of Galactic environments: Perseus and the Galactic plane region around l=11{deg} (HiGal-11). We post-processed the combined dust continuum emission images to generate column density and temperature maps. We compared these to previously adopted constant-offset corrections. We find significant differences (>~20%) over significant (~15%) areas of the maps, at low column densities (N_H_<~10^22^cm^-2^) and relatively high temperatures (T>~20K). We have also applied our method to synthetic observations of a simulated molecular cloud to validate our method. Our method successfully corrects the Herschel images, including both the constant-offset intensity level and the scale-dependent background variations measured by Planck. Our method improves the previous constant-offset corrections, which did not account for variations in the background emission levels.