We propose to perform a high cadence monitoring of nearby galaxies that have high probabilities of hosting supernovae (SNe). SNe light curves have been used to understand the expansion history of the universe, and a lot of efforts have gone into characterizing the overall shape of the radioactively powered light curve and using them as standardized candles for distance measurements. However, we still have little direct observational evidence for the theorized SN progenitor systems. Recent studies suggest that the early light curve of supernova shortly after its explosion contains valuable information about its progenitor system and can be used to set a limit on the progenitor size, R*. By performing 0.5 day or less interval monitoring of nearby galaxies (d < 50 Mpc) with 1-m class telescopes, we will be able to catch the very early precursor (shock-heated) emission as faint as R=21 mag or so. We will perform B and R or R-band only imaging monitoring for 1-3 hrs every night, in coordination with other telescopes in Korea, US, Australia, and Uzbekistan. With a network of these telescopes, the cadence of the monitoring can be as small as 8 hours. Such an observation will allow us to catch the very early light curve of about 10 SNe over three year (statistically largest sample of this kind), provide stringent constraints on their progenitor system, and enhance our knowledge on the last stage of the stellar evolution. The close distances to the host galaxies and the moderately deep observation ensure that we reach sensitivity deeper than other similar studies (e.g., K2 survey of the Kepler).