If you have a set of drone imagery and you want to know the location and direction of the camera, this tutorial might be interesting for you. We have done an applied example that retrieves the geospatial metadata from the drone camera for a group of images and makes a geostapatial repretacion on a map with the image name available as popup. The tutorial is done under a Jupyter notebook with Python and Folium.

For this tutorial you need a geospatial conda environment. Please follow this other tutorial:

hatarilabs.com/ih-en/how-to-install-python-geospatial-libraries-gdal-fiona-rasterio-etc-under-a-conda-env-in-windows

Tutorial

Code

#!pip install exif
#!pip install folium
import os
import folium
import numpy as np
import pandas as pd
from exif import Image

imageFolder = '../VUELO1/'
imageList = os.listdir(imageFolder)
#filter list to image files
imageList = [i for i in imageList if i[-4:]=='.JPG']
imageList[:5]

['DJI_0591.JPG',
 'DJI_0592.JPG',
 'DJI_0593.JPG',
 'DJI_0594.JPG',
 'DJI_0595.JPG']

For one image

imagePath = os.path.join(imageFolder, imageList[0])
imageRead = open(imagePath,'rb')
imageObject = Image(imageRead)
imageObject.list_all()[:5]

['image_description', 'make', 'model', 'orientation', 'x_resolution']

#show geospatial exif records
print(imageObject.gps_latitude)
print(imageObject.gps_longitude)
print(imageObject.gps_latitude_ref)
print(imageObject.gps_longitude_ref)

(15.0, 27.0, 59.0756)
(74.0, 42.0, 40.3897)
S
W

def getCoords(imageObject):
    latTuple = imageObject.gps_latitude
    lonTuple = imageObject.gps_longitude
    latRef = imageObject.gps_latitude_ref
    lonRef = imageObject.gps_longitude_ref
    latCoord = latTuple[0] + latTuple[1]/60 + latTuple[2]/3600
    lonCoord = lonTuple[0] + lonTuple[1]/60 + lonTuple[2]/3600
    #for negatives lats and lons
    if latRef == 'S':
        latCoord = -latCoord
    if lonRef == 'W':
        lonCoord = -lonCoord
    return latCoord, lonCoord

#get lat and lon for one image
lat, lon = getCoords(imageObject)
print(lat, lon)

-15.466409888888888 -74.71121936111112

Working for all images

#get an pandas dataframe for lats and lons
latLonDf = pd.DataFrame(columns=["Lat","Lon"])
for image in imageList:
    imagePath = os.path.join(imageFolder, image)
    imageRead = open(imagePath,'rb')
    imageObject = Image(imageRead)
    lat, lon = getCoords(imageObject)
    #print(lat, lon)
    latLonDf.loc[image,"Lat"]=lat
    latLonDf.loc[image,"Lon"]=lon
latLonDf.head()

### Plotting images coordinates
meanLat = latLonDf.Lat.mean()
meanLon = latLonDf.Lon.mean()
m = folium.Map(location=[meanLat,meanLon], zoom_start=17, control_scale=True)
for index, row in latLonDf.iterrows():
    folium.CircleMarker(location=[row.Lat,row.Lon], radius=2, popup=index).add_to(m)
m

Make this Notebook Trusted to load map: File -> Trust Notebook

latLonDf.to_csv('imagesLatLon')

Input data

You can download the input data from this link.

—————
TRANSCOM ISP – Free Sigma HSE Email
Level 6 Domain Names are FREE – Register Now.