We explore how well crowded field point-source photometry can be accomplished with Sloan Digital Sky Survey ( SDSS ) data .
For this purpose , we present a photometric pipeline based on DoPHOT ( 50 ) , and tuned for analyzing crowded-field images from the SDSS .
Using Monte Carlo simulations we show that the completeness of source extraction is above 80 \% to an i band AB magnitude of \lesssim 21 and a stellar surface density of \sim 200 arcmin ^ { -2 } .
Hence , a specialized data pipeline can be efficiently used for fairly crowded fields , such as nearby resolved galaxies in SDSS images , where the standard SDSS photometric package Photo , when applied in normal survey mode , gives poor results .

We apply our pipeline to an area of \sim 3.55 \Box ^ { \circ } around the dwarf spheroidal galaxy ( dSph ) Leo I .
Using the resulting multi-band ( g , r , i ) photometry we construct a high signal-to-noise star-count map of Leo I , utilizing an optimized filter in color-magnitude space .
This filter reduces the foreground contamination by \sim 80 \% and enhances the central stellar surface density contrast of the dwarf by a factor of \gtrsim 4 , making this study the deepest wide-field study of the Leo I dSph based on accurate CCD photometry .
We find that the projected spatial structure of Leo I is ellipsoidal .
The best fitting empirical King model to the stellar-surface density profile yields core and tidal radii of ( 6.21 \pm 0.95 ) ^ { \prime } and ( 11.70 \pm 0.87 ) ^ { \prime } , respectively .
This corresponds to ( 460 \pm 75 ) pc and ( 860 \pm 86 ) pc assuming a distance to Leo I of 254 ^ { +19 } _ { -16 } kpc .
The radial surface-density profile deviates from the King profile towards outer radii , yet we find no evidence for ’ S ’ shaped or irregular tidal debris out to a stellar surface-density of 4 \hbox { $ \times$ } 10 ^ { -3 } of the central value .
From the luminosity function of all possible Leo I stars , which we carefully extrapolated to faintest magnitudes , we determine the total I _ { c } -band luminosity of Leo I to be ( 3.0 \pm 0.3 ) \hbox { $ \times$ } 10 ^ { 6 } \mbox { $L _ { I _ { c } , \odot } $ } .
We model the mass of the dSph using the spherical and isotropic Jeans equation and infer a central mass density of 0.07 \hbox { $M _ { \odot } $ } \mathrm { pc } ^ { -3 } leading to a central mass-to-light ratio of \sim 3 in I _ { c } band solar units .
Assuming that the mass in the system follows the distribution of the visible component , we constrain a lower limit on the total mass of the dSph to be ( 1.7 \pm 0.2 ) \hbox { $ \times$ } 10 ^ { 7 } \hbox { $M _ { \odot } $ } .
On the other hand , if the mass in Leo I is dominated by a dark-matter ( DM ) halo with constant density , then the mass within the central 12 ^ { \prime } yields ( 2 \pm 0.6 ) \hbox { $ \times$ } 10 ^ { 8 } \hbox { $M _ { \odot } $ } .
Combining the inferred mass estimates with the total luminosity leads to a mass-to-light ratio of \gg 6 in I _ { c } band solar units , and possibly > 75 if the DM halo dominates the mass and extends further out than 12 ^ { \prime } .
In summary , our results show that Leo I is a symmetric , relaxed and bound system ; this supports the idea that Leo I is a dark-matter dominated system .