The positioning accuracy of the footprint of a satellite laser altimeter is primarily dependent on the accuracy of its laser pointing, e.g., a 30 arcsec pointing bias will induce 87 m horizontal error and 1.5 m vertical error when the altitude is 600 km and the laser incident angle is 1. In order to achieve the three-dimensional high-precision observation on the Earth surface, on-orbit calibration is needed to remove the systematic pointing bias mainly arising from the thermal effect. The current methods of on-orbit calibration and verification for laser altimeters are the attitude maneuvering and the footprint detection, respectively. However, the attitude maneuvering is not applicable to the existing satellite platform of China, which uses the large platform with a three-axis attitude stabilization system. The current footprint detection method can only achieve on-orbit verification task, i.e., the horizontal and vertical errors can be evaluated by analyzing the captured laser footprints but the systematic pointing bias cannot be estimated and removed. An improved design scenario of energy detector that is used for capturing laser footprint is given in this paper. The quantification level of the captured laser energy is equal to 8, which is bigger than that of the energy detector designed for geoscience laser altimeter systems corresponding to level 2. Benefiting from the new design scenario, fewer detectors are needed to achieve the same precision when calculating the centroid geolocations of captured footprints. A new systematic misalignment estimation model in the laser direction cosines is deduced, and it is used to estimate the systematic bias by using the detected footprints based on the Gauss-Markoff criterion. With the new detectors and bias estimation model, the footprint detection method now can achieve on-orbit calibration, as well as on-orbit verification. According to the presented calculation model, simulation experiments are operated to analyse three effects that may influence the performance of the footprint detection on-orbit calibration, i.e., the laser incident angle on the detector array, the surface roughness of the site where detectors lay out, and the grid density of the detector array. The simulation results indicate that, when the horizontal positioning accuracy of the captured footprint centroid demands better than 1.8 m which corresponds to 0.6 arcsec laser pointing accuracy when the altitude of the satellite is 600 km, the grid distance of the detector array can be 20 m, the laser incident angle on the detector array should be larger than 3, and the surface roughness of the calibration site should be less than 0.1 m. The designed detectors and calibration method will be used to capture laser footprints and remove the systematic bias for the laser altimeter on China GF-7 satellite, which is one of the upcoming high-resolution satellites for Earth observation.