Drone Data Collection & Processing

Remote sensing is the science and technology of obtaining information about an object, surface, or phenomenon from a distance without making physical contact with it. In aerial remote sensing, the data is collected using elevated platforms such as:

• Drones (UAVs)

• Aircraft

• Satellites

The sensors mounted on these platforms detect and measure reflected or emitted energy across different wavelengths of the electromagnetic spectrum. This information is then processed to analyze the characteristics of the target area. Aerial remote sensing is widely used in:

• Agriculture

• Surveying and mapping

• Environmental monitoring

• Urban planning

• Disaster management

• Weather and climate studies

Aerial data refers to photographs, images, measurements, or sensor-based information captured from elevated platforms such as drones, aircraft, or satellites. Unlike normal ground-level photography, aerial data provides:

• A top-down or oblique perspective of the Earth’s surface

• Large-area spatial coverage

• High-resolution mapping capability

• Georeferenced and measurable outputs

Aerial data is essential for:

• Terrain analysis

• 2D and 3D mapping

• Land use classification

• Crop health monitoring

• Change detection

• Infrastructure inspection

Remote sensing can be broadly classified into two major categories:

1. Passive Remote Sensing

2. Active Remote Sensing

Type	Description	Examples
Passive	Detects natural radiation emitted or reflected by the object, usually sunlight.	Landsat, multispectral cameras, RGB satellite imagery
Active	Emits its own energy and measures the reflected signal to gather information.	Radar, SAR, LiDAR

Ground Control Points (GCPs) are physical markers placed on the ground with accurately measured geographic coordinates such as:

• Latitude

• Longitude

• Elevation

They serve as fixed reference points to align drone-captured images with real-world coordinates. GCPs are extremely important in improving the absolute accuracy of drone-generated maps and models. Why GCPs are important:

• Improve geospatial accuracy

• Reduce positional error

• Strengthen image alignment

• Improve orthomosaic and DEM accuracy

• Enable centimetre-level precision

Applications:

• Surveying

• Construction monitoring

• Precision agriculture

• Corridor mapping

• Mine and terrain modelling

Georeferencing is the process of assigning real-world coordinates to drone images, maps, or models so that they accurately align with geographic locations on Earth. In drone data workflows, georeferencing is achieved through:

• GPS / GNSS coordinates

• Ground Control Points (GCPs)

• RTK / PPK positioning systems

• Flight metadata and onboard navigation systems

Purpose of georeferencing:

• Convert raw aerial imagery into usable spatial data

• Enable map overlay and GIS integration

• Improve positional accuracy

• Support engineering and planning applications

Outputs:

• Orthomosaic maps

• Georeferenced point clouds

• Georeferenced DEMs / DSMs

• Accurate 2D and 3D survey products

Terrain modelling is the process of representing the Earth’s surface in digital form using drone-acquired data. This is one of the most important outputs of UAV surveying and remote sensing workflows. Major terrain modelling products include:

• Digital Elevation Model (DEM)

• Digital Surface Model (DSM)

• Digital Terrain Model (DTM)

• Contour maps

• Slope and aspect maps

Terrain modelling helps in:

• Landform analysis

• Earthwork estimation

• Flood modelling

• Route planning

• Infrastructure development

• Mining and construction projects

Common technologies used:

• UAV photogrammetry

• LiDAR scanning

• Stereo image reconstruction

Image processing is the computational workflow used to convert raw drone images into meaningful geospatial outputs. This process generally includes:

• Image alignment

• Feature matching

• Camera calibration

• Point cloud generation

• Mesh and surface creation

• Orthomosaic generation

• Radiometric correction

• Vegetation or thermal analysis

The goal of image processing is to transform raw imagery into:

• Accurate maps

• 3D models

• Terrain products

• Analytical remote sensing outputs

Common techniques:

• Structure from Motion (SfM)

• Photogrammetric reconstruction

• Object detection and classification

• Image stitching and mosaicking

Cloud computing has become increasingly important in drone remote sensing because aerial datasets are often very large and computationally intensive. Cloud-based drone processing enables:

• Online image upload and storage

• Remote photogrammetry processing

• Scalable computing for large datasets

• Faster team collaboration

• Web-based sharing of maps and models

• Integration with GIS and AI tools

Advantages of cloud-based processing:

• Reduced dependency on high-end local computers

• Better accessibility and collaboration

• Faster deployment for industry workflows

• Easy visualization and reporting

Applications:

• Construction site progress monitoring

• Agriculture dashboards

• Utility inspection platforms

• Smart city planning systems

Several free and professional remote sensing portals provide satellite and Earth observation datasets for research, analysis, and drone-based comparative studies. Major remote sensing data platforms include:

1. USGS Global Visualization Viewer (GloVis)

2. NASA Earth Observation (NEO)

3. USGS Earth Explorer

4. ESA Sentinel Data

5. NASA Earth Data

6. NOAA CLASS

7. NOAA Digital Coast

8. IPPMUS Terra

9. LANCE

10. VITO Vision

These platforms provide access to:

• Satellite imagery

• Elevation datasets

• Atmospheric data

• Oceanographic observations

• Climate and land cover products

Drone data processing requires specialized software for photogrammetry, LiDAR, terrain reconstruction, thermal analytics, and mapping workflows. The following are major software platforms used in UAV remote sensing:

A typical drone-based remote sensing workflow includes the following stages:

1. Mission Planning

• Define area of interest

• Select altitude, overlap, and sensor type

2. Data Acquisition

• Fly drone mission

• Capture RGB, multispectral, thermal, or LiDAR data

3. Ground Reference Setup

• Place GCPs or use RTK / PPK for high accuracy

4. Data Processing

• Align images

• Generate point cloud, DEM, DSM, orthomosaic

5. Analysis

• Perform vegetation, terrain, change, or thermal analysis

6. Delivery

• Export maps, reports, models, dashboards, and GIS layers

!pip install face_recognition opencv-python matplotlib

Downloading face_recognition_models-0.3.0.tar.gz (100.1 MB)

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 100.1/100.1 MB 7.8 MB/s eta 0:00:00

It will continue until it shows

Successfully built face-recognition-models Installing collected packages: face-recognition-models, face_recognition Successfully installed face-recognition-models-0.3.0 face_recognition-1.3.0

from IPython.display import display, Javascript

from google.colab.output import eval_js

import numpy as np

import cv2

import base64

def take_photo(filename='photo.jpg', quality=0.8):

js = Javascript('''

async function takePhoto(quality) {

const div = document.createElement('div');

const capture = document.createElement('button');

capture.textContent = 'Capture';

div.appendChild(capture);

const video = document.createElement('video');

video.style.display = 'block';

const stream = await navigator.mediaDevices.getUserMedia({video: true});

document.body.appendChild(div);

div.appendChild(video);

video.srcObject = stream;

await video.play();

await new Promise((resolve) => capture.onclick = resolve);

const canvas = document.createElement('canvas');

canvas.width = video.videoWidth;

canvas.height = video.videoHeight;

canvas.getContext('2d').drawImage(video, 0, 0);

stream.getVideoTracks()[0].stop();

div.remove();

return canvas.toDataURL('image/jpeg', quality);

}

''')

display(js)

data = eval_js(f'takePhoto({quality})')

image_bytes = base64.b64decode(data.split(',')[1])

np_arr = np.frombuffer(image_bytes, np.uint8)

img = cv2.imdecode(np_arr, cv2.IMREAD_COLOR)

cv2.imwrite(filename, img)

return img

import face_recognition print("Capture KNOWN face")

known_frame = take_photo('known.jpg')

known_image = face_recognition.load_image_file('known.jpg')

known_encoding = face_recognition.face_encodings(

known_image,

num_jitters=50,

model='large'

)[0]

print("Capture TEST face")

test_frame = take_photo('test.jpg')

import matplotlib.pyplot as plt

face_locations = face_recognition.face_locations(test_frame)

face_encodings = face_recognition.face_encodings(

test_frame,

face_locations,

num_jitters=23,

model='large'

) for (top, right, bottom, left), face_encoding in zip(face_locations, face_encodings):

match = face_recognition.compare_faces([known_encoding], face_encoding)[0]

label = "Recognized" if match else "Unrecognized"

color = (0, 255, 0) if match else (0, 0, 255)

cv2.rectangle(test_frame, (left, top), (right, bottom), color, 2)

cv2.putText(test_frame, label, (left, top - 10),

cv2.FONT_HERSHEY_SIMPLEX, 0.9, color, 2)

print("Enter..." if match else "Unrecognized")

plt.imshow(cv2.cvtColor(test_frame, cv2.COLOR_BGR2RGB))

plt.axis('off')

plt.show()

1. S.S. Chin. Missile Configuration Design. McGraw Hill, 1961.

2. Mark Pinney Aerodynamics of Missiles and Rockets. McGraw-Hill Education, 2013.

3. Marvin Hobbs Fundamentals of Rockets, Missiles, and Spacecraft. J.F. Rider, 1962.

4. Sethunathan, P., Sugendran, R. N., & Anbarasan, T. Aerodynamic Configuration design of a missile at Int J Eng Res & Technol (IJERT), 2015.

5. Jack N. Nielsen Missile Aerodynamics. NIELSEN ENGINEERING & RESEARCH, INC, 1988.

6. Siouris, George Missile Guidance and Control Systems. Springer New York, 2006.