Longwall mining remains one of the most efficient methods for underground coal recovery. A key aspect in achieving safe and productive longwall operations relies on maintaining the shearer in an optimal position for extraction within the coal seam. The typical approach to this resource identification issue is labour intensive so is subject to safety and productivity drawbacks. As a solution, this paper describes the use of thermal infrared-based sensing to provide a means to automatically measure the vertical position of the mining machine with respect to the coal seam. This is achieved by identifying and tracking non-optically visible horizontal line-like bands in the main body of coal, which are known as marker bands. These marker bands are strongly linked to the profile of coal seam structure, a geological characteristic often used by operators as an ad hoc datum for maintaining in-seam alignment of the shearer. Details on the theory behind thermal infrared imaging and practical aspects involved in implementation of the method are given. As there are very few real-time solutions available to locate and track coal seam profiles, this approach overcomes a current limitation in implementing intelligent horizon control systems for advanced shearer operation. Measurements from a shearer-based sensing system are given to demonstrate the approach.