Efficient collection and redistribution of the direct and diffuse components of daylight in buildings remains a major objective of advanced fenestration systems. Such systems, including novel solar blinds, new glazing or coating materials and daylight-redirecting devices, can improve significantly the penetration of daylight in deep rooms to reduce electricity consumption while improving visual comfort conditions greatly; at the same time, they can lead to larger solar gains in winter combined with lower solar loads in summer. To allow their integration in buildings and benefit from their potential as energy-efficient strategies, an in-depth and accurate knowledge of their directional optical properties is necessary. These properties are described by Bidirectional Transmission (or Reflection) Distribution Functions, abbreviated BT(R)DF, that express the emerging light flux distribution for a given incident direction. Such detailed transmission or reflection functions are intended to be used by the building industry to optimize the luminous performances of innovative solutions for windows and to describe photometric properties of complex glazing and shading materials according to a common format. On the other hand, they will allow daylighting simulation tools to improve their potentialities and integrate complex fenestration systems reliably in the simulation models. Their accurate assessment requires an appropriate experimental equipment. An innovative bidirectional goniophotometer, based on digital imaging techniques, was designed and set up for that purpose: instead of scanning the emerging light flux distribution by moving a sensor from point to point, an original method was used that comprises a rotating diffusing screen on which the emerging light flux is reflected towards a digital video-camera, used as a multiple-points luminance-meter. This novel approach significantly reduces the time needed to monitor BT(R)DF data, lowering it down to a few minutes per incident direction instead of several hours for conventional assessment methods, which is a critical parameter in BT(R)DF assessment as about a hundred incident directions are usually required. At the same time, it allows the gathering of continuous transmitted (reflected) light distribution figures, only limited in resolution by the pixellisation of the digital images. Moreover, by taking advantage of the considerable luminance range reached when images are captured and superposed at different integration intervals, combined to the appreciable flexibility in the data processing offered by digital imaging-based techniques, a remarkable accuracy can be achieved when assessing BT(R)DF data. This PhD thesis explains the different conception, calibration and processing stages that were necessary to develop the bidirectional goniophotometer into a functioning, validated measurement device. Its design for combining BTDF and BRDF assessments is described, the various calibration procedures for converting the CCD camera into a reliable multiple-points luminance-meter are detailed, as well as the image and data processing phases. An in-depth validation of the performed measurements was realized based on different approaches and led to a relative error on BT(R)DF data of only 10%, allowing to confirm the high accuracy and reliability of this novel device.