The fractions of four components （sunlit leaves

shaded leaves

sunlit soil and shaded soil） and gap fraction are characteristic geometric parameters of canopy’s structure

and have great influence on radiative transfer of the vegetation. In this paper

POV-ray software was introduced to efficiently and accurately calculate the four components’ fractions and gap fraction of three-dimensional （3D） canopies. With the good agreements of computed four components’ fractions and gap fraction be- tween POV-ray and Radiosity-Graphics combined Model （RGM） based on the same 3D canopy structure

more 3D structures generated by the Clumped Architecture Model of Plants （CLAMP） model were used to study the characteristics of four compo- nents’ fractions and gap fraction. The key parameters of CLAMP model include sowing scheme

clumping index

average leaf angle （ALA） and leaf area index （LAI）. By analyzing the sensitivity of these parameters to four components’ fractions and gap fraction

the changing features of four components’ fractions and gap fraction were found. The results show that: clumping index can effectively control the spatial distribution of leaves

which affects the relative proportion of vegetation and soil in the scene. Therefore

clumping index has a great impact on the distribution pattern of four components’ fractions and gap fraction. Average leaf angle influences the anisotropy of the results. When the average leaf angle is larger

the magnitude of anisotropy is higher. Compared with other parameters

the impact of row effect on the four components’ fractions and gap fraction is primary when the scene’s row structure is obvious. Therefore

when the canopy is in a row structure

the characteristic change of four components’ fractions and gap fraction is more complex than that of homogeneous canopy. At the same time

the results of row canopy in three typical growing stages were compared

which shows that soil has a higher contribution to the four components’ fractions and gap fraction than that of vegetation at the early stage. With the increase of leaf area index

the influence of vegetation on the result is gradually enhanced and the fraction of sunlit leaves shows more apparent anisotropy. The research of this paper shows that POV-ray is a powerful tool to quantitatively study vegetation by remote sensing technology

and has potential of wider applications.