I present J- and Ks-band galaxy counts and galaxy colors covering 750<br />arcminutes<sup>2</sup> in the deep AKARI North Ecliptic Pole (NEP) field, using the<br />FLeridA Multi-object lmaglng Near-ir Grism Observational Spectrometer<br />(FLAMINGOS) on the Kitt Peak National Observatory (KPNO) 2.1 m<br />telescope. The limiting magnitudes with a slgnal-to-noise ratio of three in<br />the deepest regions are 21.85 and 20.15 in the J- and Ks-bands respectively<br />in the Vega magnitude system.<br /><br /> The J-and Ks-band galaxy counts in the AKARINEP field are broadly in<br />good agreement with those of other results in the literature, however the<br />counts show a characteristic change in the galaxy count slope in both the J-and<br />Ks-bands. The slope of the J-band galaxy counts per square degree per<br />magnitude clearly changes at J ~ 19.5 from d(logN)/dm = 0.39 ± 0.02 to<br />0.30 ± 0.03, while the ks-band galaxy counts shows a clear 'bump' in<br />the magnitude range 18.0 < Ks < 19.5. The B -Ks color distribution of the<br />galaxy samples also shows an obvious change around the `bump'<br />magnitude of the ks-band counts. The number of faint blue samples in the<br />magnitude range 18.5 <Ks < 19.5 is signincantly larger than that of ks<<br />17.5 samples.<br /><br /> In order to find a clue tounderstandthe ongln of the change in J- and<br />ks-band count slopes and the bluer trend at fainter magnitudes below the<br />bump of Ks-band counts, I also investigated the number counts of Ks-band<br />selected high redshift galaxy populations:the extremely red objects (EROs),<br />the B,z and K-band selected galaxies (BzKs) and the distant red galaxies<br />(DRGs). I foundthat the number counts of both the passive EROs and the<br />passive BzKs have shallower slopes at fainter magnitudes and turnovers at<br />Ks ~ 18.5 and Ks ~ 19.0, respectively, while the number count of the<br />star-fomlng BzKs is a single power-law with a steeper slope. The observed<br />number counts of these passive populations are compared with a backward<br />evolution model allowing the number density evolution. The pure<br />luminosity evolution model without density evolution fails to explain the<br />observed counts at faint magnitudes (Ks > 17.0), while the model with a<br />negative density evolution is consistent with the observed counts of the<br />passive populations.<br /><br /> Based on these results, I also investigated an evolutional scenario of<br />galaxy populations which can explainthe observed K-band count by using<br />three components (early-type, late-type and star-forming populations),<br />backward evolution model with type-dependent luminosity functions. I<br />found that not onlythe ellipticals but also the spiral galaxy population has<br />to follow the negative density evolution whilethe starburst population<br />shows the strong positive density evolution.<br /><br /> Finally,the on-going NEP deep surveys at 2-26 pm with AKARI will<br />provide us with new and critical pieces of information to verify and improve<br />the galaxy evolution scenario discussed in this thesis.