diff --git a/help/Functions_description.md b/help/Functions_description.md
index 0fdbe18..f6b3738 100644
--- a/help/Functions_description.md
+++ b/help/Functions_description.md
@@ -247,7 +247,7 @@ Compact-pol functionalities require the SAR data in the form of 2x2 covariance m
- * Compact-pol Radar Vegetation Index (CpRVI) This functionality computes the compact-pol radar vegetation index for compact polarimetric SAR data. The required input and the computed output are as follows:
+ * CpRVI (Compact-pol Radar Vegetation Index): This functionality computes the compact-pol radar vegetation index for compact polarimetric SAR data. The required input and the computed output are as follows:
````python
input : input_C2_folder, window_size
@@ -317,7 +317,7 @@ Compact-pol functionalities require the SAR data in the form of 2x2 covariance m
- * DOP (Degree of Polarization) This functionality computes the degree of polarization for compact polarimetric SAR data. The required input and the computed output are as follows:
+ * DOP (Degree of Polarization): This functionality computes the degree of polarization for compact polarimetric SAR data. The required input and the computed output are as follows:
````python
@@ -383,58 +383,78 @@ Dual-pol functionalities require the SAR data in the form of 2x2 covariance matr
- * Dual-pol Radar Vegetation Index (DpRVI)
+ * DpRVI (Dual-pol Radar Vegetation Index): This functionality computes the dual polarimetric radar vegetation index for dual polarimetric (HH | HV), (VV | VH) SAR data. The required input and the computed output are as follows:
+
+ ````python
+ input : input_C2_folder, window_size
+ output: DpRVI.bin
+ ````
+
+ The formulation of DpRVI is as follows:
+
- ![grvi](https://latex.codecogs.com/svg.latex?\Large&space;\text{DpRVI}=1-\Big(\frac{\lambda_1}{\lambda_1+\lambda_2}\Big)\sqrt{1-\frac{4\times\text{det([C2])}}{\text{(Trace[C2])}^2}})
+ ![dprvi](https://latex.codecogs.com/svg.latex?\Large&space;\text{DpRVI}=1-\Big(\frac{\lambda_1}{\lambda_1+\lambda_2}\Big)\sqrt{1-\frac{4\times\text{det([C2])}}{\text{(Trace[C2])}^2}})
-where, C2 is co-variance matrix, and λ1 and λ2 are the eigen values of C2 matrix in descending order.
+ where, C2 is co-variance matrix, and λ1 and λ2 are the eigen values of C2 matrix in descending order (λ1 > λ2). Further details on DpRVI can be obtained from [[5]](#5)
-````python
- input : input_C2_folder, window_size
- output: DpRVI.bin
-````
- * Radar Vegetation Index (RVIdp)
-
-![grvi](https://latex.codecogs.com/svg.latex?\Large&space;\text{RVI}_{dp}=\frac{4\sigma^\circ_{\text{XY}}}{\sigma^\circ_{\text{XX}}+\sigma^\circ_{\text{XY}}})
-
-
+ * RVI (Radar Vegetation Index): This functionality computes the radar vegetation index for dual polarimetric (HH | HV), (VV | VH) SAR data. The required input and the computed output are as follows:
+
+ ````python
+ input : input_c2_folder, window_size
+ output: RVI_dp.bin
+ ````
+ The formulatin of RVI is as follows:
-````python
- input : input_c2_folder, window_size
- output: RVI_dp.bin
-````
+
- * Polarimetric Radar Vegetation Index (PRVIdp)
-
+ ![rvidp](https://latex.codecogs.com/svg.latex?\Large&space;\text{RVI}_{dp}=\frac{4\sigma^\circ_{\text{XY}}}{\sigma^\circ_{\text{XX}}+\sigma^\circ_{\text{XY}}})
+
+
+
+ where, σoXX and σoXY co- and cross-pol backscatter intensities.
-![grvi](https://latex.codecogs.com/svg.latex?\Large&space;\text{PRVI}_{dp}=(1-\text{DOP}_{dp})\sigma^\circ_{\text{XY}})
+ * PRVI (Polarimetric Radar Vegetation Index): This functionality computes the polarimetric radar vegetation index for dual polarimetric (HH | HV), (VV | VH) SAR data. The required input and the computed output are as follows:
+
+ ````python
+ input : input_c2_folder, window_size
+ output: PRVI_dp.bin
+ ````
-
+ The formulation of PRVI is as follows:
+
+
+
+ ![prvidp](https://latex.codecogs.com/svg.latex?\Large&space;\text{PRVI}_{dp}=\Big(1-\sqrt{1-\frac{4\times\text{det([C2])}}{\text{(Trace[C2])}^2}}\Big)\sigma^\circ_{\text{XY}})
+
+
-````python
- input : input_c2_folder, window_size
- output: PRVI_dp.bin
-````
- * Degree of Polarization (DOPdp)
-
-![grvi](https://latex.codecogs.com/svg.latex?\Large&space;\text{DOP}_{dp}=\sqrt{1-\frac{4\times\text{det([C2])}}{\text{(Trace[C2])}^2}})
+
+ * DOP (Degree of Polarization): This functionality computes the 2D Barakat degree of polarization for dual polarimetric (HH | HV), (VV | VH) SAR data. The required input and the computed output are as follows:
-
+ ````python
+ input : input_c2_folder, window_size
+ output: dop_dp.bin
+ ````
+
+
+
+ ![dopdp](https://latex.codecogs.com/svg.latex?\Large&space;\text{DOP}_{dp}=\sqrt{1-\frac{4\times\text{det([C2])}}{\text{(Trace[C2])}^2}})
+
+
+
+ Further details on the Barakat Degree of polarization can be found in [[10]](#10)
+
-````python
- input : input_c2_folder, window_size
- output: dop_dp.bin
-````
## References