How to calculate percentage of white portion of an image to measure the density of the vehicles ?

I have tried one program. I m getting the white portion of an image in pixels. But i want it in percentage.

And I wanted to calculate the density of the vehicles on the road. Wanted to remove the other areas like trees, divider of the road etc.

Image is attached herewith.

Program is:

clc;
clear;
close all;
a=imread('frame1.jpg');% read RGB image (frame no. 1)
imshow(a);
%Selection of ROI
[r c]=ginput(4);
BW=roipoly(a,r,c);
figure;imshow(BW);
[R C]=size(BW);
for i=1:R
    for j=1:C
        if BW(i,j)==1
            Out(i,j)=a(i,j);
        else
            Out(i,j)=0;
        end
    end
end
figure;imshow(Out,[]);title('Output Image');
 
%Convert ROI Selecetd image to binary image
threshold = 165; % threshold set at 165
z= Out>threshold; % convert grayscale image to binary image
figure;imshow(z)% display binary image
 
% Dilation
y2 = bwmorph(z,'dilate',1) ;
figure ; imshow(y2)
 
%Connected component labelling
bwlabel(y2,4)% containing labels for the connected objects in y2, where 4 specifies 4-connected objects
 
 
%Blob analysis
 
%Filling holes
C=imfill(y2,'holes'); 
figure;imshow(C);
 
 
%Apply Median filter to remove Noise
FilteredImage=medfilt2(C,[5 5]);
figure;imshow(FilteredImage)
 
%Boundary Label the Filtered Image
[L num]=bwlabel(FilteredImage);%returns in num the number of connected objects found in FilteredImage
STATS=regionprops(L,'Area');
cc=[];
removed=0;
 
%Remove the noisy regions 
for i=1:num
dd=STATS(i).Area;
if (dd < 50)% when we change the value as 20, 50, 100,500,5000 no. of blobs will change 
L(L==i)=0;
removed = removed + 1;
num=num-1;
else
end
end
[L2 num2]=bwlabel(L);
 
% Trace region boundaries in a binary image.
[B,L,N,A] = bwboundaries(L2);
 
%Display results
subplot(1,2,1),  imshow(L2);
title('BackGround Detected');
subplot(1,2,2),  imshow(L2);
title('blob detected');
hold on;
 
for k=1:length(B),
if(~sum(A(k,:)))
boundary = B{k};
plot(boundary(:,2), boundary(:,1),'r','LineWidth',2);
end
end
 
%find area of each blob
%labeledImage = bwlabel(L2); %Label connected components in image L2
%measurements = regionprops(labeledImage, 'Area');%measures the property "area"in the labeled image
%allAreas = [measurements.Area];  % List of all the blob areas.
%totalAreaOfAllBlobs = sum(allAreas); % Will be the same as [pixelSum2 = sum(binaryImage(:)]
 
 
% calculate white portion of the image%
nblack=sum(L2(:));
nwhite=numel(L2)-nblack;