To date, powerful lasers such as argon or ruby lasers have been used in most applications of particle image velocimetry. We present an analysis of the ability of this technique when a low power (5 mW He-Ne) laser is used. As a practical example we apply it to the Rayleigh-Benard convective flow, where velocities ranging up to several mm/s have been measured with about 100µJ energy pulses. The parameters that allow us to obtain good quality photographs with such low light power are analyzed. Several examples of pictures taken from different flow planes at different Rayleigh numbers are shown. The data reduction is made by fringe analysis. The digital image of the Young's fringes is recorded with a videocamera at each point of the photograph and is then processed with the 1-D averaging technique. This technique is the fastest but has the drawback of needing an external operator to select the fringe direction. To avoid this problem, we implement an algorithm whose practical performance is analyzed here. Another improvement proposed for this technique is the pedestal removal from the signal by digital high pass filtering. In this way, automatic detection of the first peak in the Fourier transform of the 1-D averaged signal is reached more easily. Thus, full automation of the 1-D averaging technique is obtained. The spatial filtering technique is also used. Several examples of whole-field velocity maps obtained from the photographs are shown, allowing a quick analysis of the velocity field. Finally, results obtained from postprocessing of the velocity measurements are shown. They include velocity, vorticity, and streamline maps. Isovelocity contour maps are also generated with the computer from the velocity field to compare with the maps obtained from the spatial filtering technique.