Empirical_Calibration.m

%% Find Files & Put in Metadata

% This code allows you to calibrate a Snow Depth pole. It assumes you're
% working with a snow depth pole that has tick marks every 5 centimeters

clear all; close all; clc % fresh start
set(0,'DefaultFigurePosition',[0          41*40        658       658]); % changes the size of the window

% You'll need to update these paths based on where you've stored your images the first addpath should simply just add the directory, while the second should include the wildcard *.JPG at the end.
addpath('/Users/wcurrier/Desktop/Matlab_Functions') % you may need some functions but I don't think so
addpath('/Volumes/Seagate_Backup_Plus_Drive/OLYMPEX_Photos/Anderson_Pass/camera_1/Pole_2'); % path to where all your photos are
listing = dir('/Volumes/Seagate_Backup_Plus_Drive/OLYMPEX_Photos/Anderson_Pass/camera_1/Pole_2/*.JPG'); % path to where

numImgs = size(listing,1); % determines the number of images in the directory you pointed to


site_name='Anderson_Pass_West';
Declination_Angle='15 degrees';
color='Green';
Height_of_Camera=6.19; % meters
Distance_to_Pole=19.6; % cm
Elevation=4600; %feet
Lat=47.69773333333333; % decimal degrees N
Long=-123.32686666666666; % decimal degrees W
camera_number=1;
pole_number=2;
pole_length=500; % centimeters
NumberOfTicks=pole_length/5; % Number of ticks - 400 centimeter pole should have 80 Tick Marks (every 5 cm)
Image_Number=[196 1019 1035]; % Based on the listing what is the image number that you want to calibrate to
Image_name=listing(Image_Number).name;
%% Code Starts Here
for ii=1:length(Image_Number)
    close all
    
 % Store the name of the image you used to calibrate

    Image_name=listing(Image_Number).name; % Store the name of the image you used to calibrate
                originalImage = imread(listing(Image_Number(ii)).name); % read in Calibration Image
                RotatedImage = imrotate(originalImage,90); % Rotate Image so that you can make it larger on your screen
                imshow(RotatedImage); % show image
                [X,Y]         = ginput(4); % crop the image by clicking around the pole
                cropped_RotatedImage=RotatedImage(min(Y):max(Y),min(X):max(X),:);
                    
                    imshow(cropped_RotatedImage,'InitialMagnification', 400) % show cropped image
                    [ClickCoordinates]         = ginput(NumberOfTicks+1); % click from left to right down the pole at each tick mark
                    
                    % Store the Coordinates corrected for the non-cropped image
                    ClickCoordinates2(:,1)=min(X)+ClickCoordinates(:,1); % get the coordinates so they make sense based on the original image
                    ClickCoordinates2(:,2)=min(Y)+ClickCoordinates(:,2); % get the coordinates so they make sense based on the original image

                    %View Calibration
                    imshow(RotatedImage),hold on
                    plot(ClickCoordinates2(:,1),ClickCoordinates2(:,2),'o')
                    pause()
                    hold off              
%% Calculate Distances and Creat Look Up Table
Distances=nan(length(ClickCoordinates2),1);
for kk=1:length(ClickCoordinates2) % Use the Distance Formula To Calculate Distances
    Distances(kk,1)=sqrt((ClickCoordinates2(1,1)-ClickCoordinates2(kk,1))^2+(ClickCoordinates2(1,2)-ClickCoordinates2(kk,2))^2);
end

% Create Look Up Table

% Original Lookup Table Based on Clicks - this includes the coordinates of
% where you click
    originaltable=[[0:5:pole_length].', ClickCoordinates2, Distances]; 
% Linearly Interpolate the Lookup Table to make it larger
    interpolated_pixel_lengths=interp1(originaltable(2:end,1),originaltable(2:end,4),5:0.5:pole_length).';
    x=5:0.5:pole_length; % create a vector from 5 through the end of the pole length in increments of 0.5
    table=nan(length(interpolated_pixel_lengths)+1,4); % initialize a new table
    table(1,1:4)=originaltable(1,1:4); % Put the 0 length information back into the new table
    
    % Skip to 5 cm and then take the interpolated pixel lengths from after 5 cm and put that in the look up table matrix
        table(2:end,1:4)=[x.', nan(length(5:0.5:pole_length),2), interpolated_pixel_lengths];
    % Add NaN's to the end of the table so that any erroneous values or
    % images where you cannot see the 
    table(end+1,1:4)=[NaN,NaN,NaN,originaltable(end,4)+(3*((originaltable(end,4)-originaltable(end-1,4))))];

%% Save Data

save(strcat('Empirical_Calibration_for_',site_name,'_Camera_',num2str(camera_number),'_Pole_',num2str(pole_number),num2str(ii)));
end