Continuous and accurate ground measurements of the fraction of absorbed (fAPAR) or intercepted (fIPAR) photosynthetically active radiation by green canopy components is important to monitor canopy functioning. fAPAR and fIPAR are sensitive to illumination conditions and non-green components during the senescence stage. While several methods have been developed to estimate fAPAR or fIPAR in the field from different methods including AccuPAR, LAI-2200 and Digital Hemispheric Photograph Photography (DHP), the differences among these methods still need more investigations. The principles on which they are based are first reviewed with due attention to the assumptions used and approximations made. Two field campaigns conducted in 2012 and 2013 in northeastern China over paddy rice fields were then used to compare fAPAR and fIPAR measured using AccuPAR, DHP and LAI-2200. Results demonstrated that considering only canopy light transmittance (fIPAR), measured with AccuPAR, DHP or LAI-2200, is a good proxy of fAPAR which is computed from AccuPAR measurements of the four fluxes of the radiation balance. However, when canopy is senescing, downward looking DHP method is recommended since it is the only method that directly measures the light intercepted by green elements. Methods based on upward looking (DHP upward, AccuPAR, LAI-2200) cannot distinguish between the green and senescent vegetation elements. Corrections based on independent measurements of the ratio of the green area index (GAI) to the plant area index (PAI) (GAI/PAI) need to be used in this case, while assuming that green and senescent elements are well mixed in the canopy volume. Downward looking DHP appears to be the preferred method for relatively short and dense canopies such as rice since it does not disturb the canopy, it is sensitive to the green elements only and allows to simulate fIPAR for any illumination conditions.