Numerous studies performed on dwarf galaxies have shown large differences in
observable properties as the morphology, color and stellar content. Beside the
well-known classes of dwarf spheroidal/dwarf elliptical systems and dwarf
Magellanic (disky) systems the so-called Blue Compact Dwarf Galaxies (hereafter
BCDGs), originally defined by Zwicky (1967, 1971) and, later
discussed by Thuan & Martin (1981), have focused attention
after the discovery by Sargent & Searle (1970) that some of
them were low metallicity systems. BCDGs show metallicity ranging from 1/6
to 1/50 solar value (extreme case being I Zw 18: Searle & Sargent
1972; Davidson & Kinman 1985; Kunth & Sargent
1986). These systems are known to host a very active stellar
formation. Scaled to the low total luminosity of these dwarf objects with
absolute magnitudes 
, this corresponds to star formation rates
able to exhaust the available gas content of the galaxy in a time much
smaller than the age of the Universe.
It is quite surprising however that only few quantitative imaging studies of the brightness distribution in BCDGs have been published insofar. Barbieri et al. (1979) performed a pioneering investigation of 15 objects on photographic plate material, suggesting that some of them could be considered as dwarf elliptical systems. Loose & Thuan (1983) have reported preliminary qualitative results on a sample of 22 BCDGs, separating 4 morphological classes based on the shape of the isophotes and the apparent distribution of the star forming knots. Loose & Thuan (1985) have also made a detailed study of Haro 2, a moderate luminosity spheroidal object housing a strong star formation activity in its center. Kunth et al. (1988), from a multicolor deep CCD imagery of a restricted sample of 7 objects, have shown that some BCDGs obey a quasi-pure r1/4 law, as do giant ellipticals, while others are assumed to be disk-like systems, since they obey an exponential law. Recently, Papaderos et al. (1996) discussed such morphological properties, using profile decomposition, for some bright BCDGs. Some objects (e.g. Pox 186) have so small apparent diameters that no firm conclusions could be derived. BCDGs thus form "a mixed bag of morphological types'', probably reflecting widely different evolutionary histories. The blue-excess starburst region is in some cases surrounded by a red envelope suggesting the presence of an evolved red giant population (Aaronson 1986; Thuan 1983). Strong color gradients might indicate, beside a change in the metallicity, the existence of old stars of population II. Other authors (Bothun et al. 1989; Salzer et al. 1991) have published images obtained in follow-up studies of active galaxies, but only a few BCDGs are studied in these papers. The case of IZw18 has attracted much more attention because of its extreme abundance characteristics and deep detailed imagery in continuum and emission lines has been published by Hua et al. (1987), Davidson et al. (1989), Dufour & Hester (1990) and Petrosian et al. (1996) while recently Hubble Space Telescope imaging allowed Hunter & Thornson (1995) to build the color-magnitude diagram of the massive star population.
Our purpose is to investigate the brightness distributions of BCDGs, their structure (presence of one or several areas of star formation, of disks, spheroids, of non-axisymmetric components, traces of possible interactions, color gradients etc...) and to search for relationships, if any exist, between these parameters and the spectroscopic indicators of galactic evolution such as the metallicity of the ionized gas or the estimation of the age of the starburst. Hence, the imaging and photometric observing program whose results are reported here is coupled with a program of detailed, high spatial resolution spectro-imaging observations that will be reported elsewhere.