The dust extinction in spiral disks can be estimated from the counts of background field galaxies, provided the deleterious effects of confusion introduced by structure in the image of the foreground spiral disk can be calibrated. Gonzalez et al. (1998) developed a method for this calibration, the ``Synthetic Field Method (SFM), and applied this concept to a HST/WFPC2 image of NGC4536. The SFM estimates the total extinction through the disk without the necessity of assumptions about the distribution of absorbers or the disk light. The poor statistics, however, result in a large error in individual measurements. We report on improvements to and automation of the Synthetic Field Method which render it suitable for application to large archival datasets. To illustrate the strengths and weaknesses of this new method, the results on NGC 1365, a SBb, and NGC 4536, a SABbc, are presented. The extinction estimate for NGC1365 is A_I = 0.6 (+0.6/-0.7) at 0.45 R_25 and for NGC4536 it is A_I = 1.6 (+1.0/-1.3) at 0.75 R_25. The results for NGC4536 are compared with those of Gonzalez et al. (1998). The automation is found to limit the maximum depth to which field galaxies can be found. Taking this into account, our results agree with those of Gonzalez et al. (1998). We conclude that this method can only give an inaccurate measure of extinction for a field covering a small solid angle. An improved measurement of disk extinction can be done by averaging the results over a series of HST fields, thereby improving the statistics. This can be achieved with the automated method, trading some completeness limit for speed. The results from this set of fields are reported in a companion paper Holwerda et al. (2005b).