The GRAPES-3 experiment at Ooty contains a large-area (560 m$^{2}$) tracking muon detector. This detector consists of 16 modules, each 35 m$^{2}$ in area, that are grouped into four supermodules of 140 m$^{2}$ each. The threshold energy of muons is $\sec(\theta)$ GeV along a direction with zenith angle $\theta$ and the angular resolution of the muon detector is $6^{\circ}$. Typically, it records $\sim 4 \times 10^{9}$ muons every day. The muon detector has been operating uninterruptedly since 2001, thus providing a high statistics record of the cosmic ray flux as a function of time over one decade. However, prior to using these data, the muon rate has to be corrected for two important atmospheric effects, namely, variations in atmospheric pressure and temperature. Because of the near equatorial location of Ooty ($11.4^{\circ}$N), the seasonal variations in the atmospheric temperature are relatively small and shall be ignored here. Due to proximity to the equator, the pressure changes at Ooty display a dominant 12 h periodic behaviour in addition to other seasonal changes. Here, we discuss various aspects of a novel method for accurate pressure measurement and subsequent corrections applied to the GRAPES-3 muon data to correct these pressure-induced variations. The pressure-corrected muon data are used to measure the profile of the solar diurnal anisotropy during 2006. The data, when divided into four segments, display significant variation both in the amplitude ($\sim 45\%$) and phase ($\sim42$ m) of the solar diurnal anisotropy during 2006, which was a period of relatively low solar activity.