This thesis aimed to improve the accuracy of radiochromic film dosimetry with a main focus on optimizing protocols and dosimetry models. The research was divided in four publications. The first paper [Mendez I, Hartman V, Hudej R, Strojnik A, and Casar B, "Gafchromic EBT2 film dosimetry in reflection mode with a novel plan-based calibration method", Med. Phys. 40, 011720 (2013)] studied film dosimetry with a flatbed scanner in reflection mode, proposed a new plan-based calibration method, and selected models for sensitometric curves and lateral corrections. In the second paper [Mendez I, Peterlin P, Hudej R, Strojnik A, and Casar B, "On multichannel film dosimetry with channel-independent perturbations", Med. Phys. 41, 011705 (2014)], different channel-independent perturbation models for radiochromic film dosimetry were analysed and compared, explaining their implicit assumptions and inherent uncertainties. Several elements of the dosimetry protocol were compared as well. The third article [Mendez I, "Model selection for radiochromic film dosimetry", Phys. Med. Biol. 60, 4089 (2015)] deepened the selection of the dosimetry protocol by examining whether lateral corrections, scanning prior to the irradiation, and multichannel methods significantly improved the accuracy and precision of film doses. Also, a general perturbation model was proposed in this article. Finally, the fourth publication [Méndez I, Šljivić Ž, Hudej R, Jenko A, and Casar B. "Grid patterns, spatial inter-scan variations and scanning reading repeatability in radiochromic film dosimetry", Physica Medica (2016)] studied several sources of uncertainty related to the repeatability of flatbed scanners. Grid patterns and positioning inaccuracies were discovered, and a new method to correct inter-scan variations was proposed.This thesis aimed to improve the accuracy of radiochromic film dosimetry with a main focus on optimizing protocols and dosimetry models. The research was divided in four publications. The first paper [Mendez I, Hartman V, Hudej R, Strojnik A, and Casar B, "Gafchromic EBT2 film dosimetry in reflection mode with a novel plan-based calibration method", Med. Phys. 40, 011720 (2013)] studied film dosimetry with a flatbed scanner in reflection mode, proposed a new plan-based calibration method, and selected models for sensitometric curves and lateral corrections. In the second paper [Mendez I, Peterlin P, Hudej R, Strojnik A, and Casar B, "On multichannel film dosimetry with channel-independent perturbations", Med. Phys. 41, 011705 (2014)], different channel-independent perturbation models for radiochromic film dosimetry were analysed and compared, explaining their implicit assumptions and inherent uncertainties. Several elements of the dosimetry protocol were compared as well. The third article [Mendez I, "Model selection for radiochromic film dosimetry", Phys. Med. Biol. 60, 4089 (2015)] deepened the selection of the dosimetry protocol by examining whether lateral corrections, scanning prior to the irradiation, and multichannel methods significantly improved the accuracy and precision of film doses. Also, a general perturbation model was proposed in this article. Finally, the fourth publication [Méndez I, Šljivić Ž, Hudej R, Jenko A, and Casar B. "Grid patterns, spatial inter-scan variations and scanning reading repeatability in radiochromic film dosimetry", Physica Medica (2016)] studied several sources of uncertainty related to the repeatability of flatbed scanners. Grid patterns and positioning inaccuracies were discovered, and a new method to correct inter-scan variations was proposed.