In this thesis, we study the degree of the canonical map of surfaces of general type. In particular, we give the first examples known in the literature of surfaces having degree d=10,11, 13, 14, 15, and 18 of the canonical map. They are presented in a self-contained and independent way from the rest of the thesis. We show also how we have discovered them. These surfaces are product-quotient surfaces. In this thesis, we study the theory of product-quotient surfaces giving also some new results and improvements. As a consequence of this, we have written and run a MAGMA script to produce a list of families of product-quotient surfaces having geometric genus three and a self-intersection of the canonical divisor large. After that, we study the canonical map of product-quotient surfaces and we apply the obtained results to the list of product-quotient surfaces just mentioned. In this way, we have discovered the examples of surfaces having degree d=10,11,14, and 18 of the canonical map. The remaining ones with degrees 13 and 15 do not satisfy the assumptions to compute the degree of the canonical map directly. Hence we have had to compute the canonical degree of these two families of product-quotient surfaces in a very explicit way through the equations of the pair of curves defining them. Another work of this thesis is the classification of all smooth surfaces of general type with geometric genus three which admits an action of a group G isomorphic to \mathbb Z_2^k and such that the quotient is a projective plane. This classification is attained through the theory of abelian covers. We obtained in total eleven families of surfaces. We compute the canonical map of all of them, finding in particular a family of surfaces with a canonical map of degree 16 not in the literature. We discuss the quotients by all subgroups of G finding several K3 surfaces with symplectic involutions. In particular, we show that six families are families of triple K3 burgers in the sense of Laterveer. Finally, in another work we study also the possible accumulation points for the slopes K^2/ \chi of unbounded sequences of minimal surfaces of general type having a degree d of the canonical map. As a new result, we construct unbounded families of minimal (product-quotient) surfaces of general type whose degree of the canonical map is 4 and such that the limits of the slopes K^2/ \chi assume countably many different values in the closed interval [6+2/3, 8].

Le but de cette thèse est de considérer différentes questions sur les groupes projectifs et sur les arrangements de droites dans le plan projectif. Un groupe projectif est un groupe qui est isomorphe au groupe fondamental d'une variété projective lisse complexe. Pour étudier les groupes projectifs, des techniques sophistiquées de topologie algébrique et de géométrie algébrique ont été développées pendant les dernières décennies, par exemple la théorie des variétés caractéristiques combinée avec la théorie de Hodge s'est montrée être un outil puissant. Les arrangements de droites dans le plan projectif ont une place centrale dans l'étude des groupes projectifs. En effet, il y a beaucoup de questions ouvertes sur les groupes projectifs, et la théorie des arrangements d'hyperplans, en particulier celle des arrangements de droites, qui est un domaine très actif de recherche, peut suggérer des solutions à ces problèmes. En outre, les problèmes sur les groupes fondamentaux de complémentaires des arrangements d'hyperplans peuvent être réduits au cas des arrangements de droites, en utilisant le bien connu Théorème de Zariski du type de Lefschetz. Assez souvent, pour étudier les groupes projectifs ou quasi-projectifs, on considère d'abord les arrangements de droites pour obtenir des idées intuitives. Dans cette thèse nous obtenons aussi des résultats d'intérêts indépendants, par exemple sur les morphismes définis sur un produit d'espaces projectifs dans le Chapitre 4, sur la fibre générale de certains morphismes dans le Chapitre 5 et les critères sur les surfaces de type générales au Chapitre 7. / The objective of this thesis is to investigate various questions about projective groups and line arrangements in the projective plane. A projective group is a group which is isomorphic to the fundamental group of a smooth complex projective variety. To study projective groups, sophisticated techniques in algebraic topology and algebraic geometry have been developed in the passed decades, for instance, the theory of cohomology jump loci, together with Hodge theory, has been proven a powerful tool. Line arrangements in the projective plane are of special interest in the study of projective groups. Indeed, there are many open questions related to projective groups, and the theory of hyperplane arrangements, and in particular that of line arrangements, which is quite an active area of research, may provide insights for these problems. Furthermore, problems concerning the fundamental groups of the complements of hyperplane arrangements can be reduced to the case of line arrangements, due to the celebrated Zariski theorem of Lefschetz type. Very often, in the study of projective groups or quasi-projective groups, one usually considers line arrangements first to get some intuitive ideas. In this thesis, we also prove some theorems that are of independent interest and can be used elsewhere, for instance, we prove properties concerning morphisms from products of projective spaces in Chapter 4, we show that some morphisms have generic connected fibers in Chapter 5 and we give criteria for a projective surface to be of general type in Chapter 7.