Post on 14-Nov-2014
description
Dept. of Geospatial & Space Technology
University of NairobiB. Sc. (Civil Engineering)
1.1. Procedures of Image ProcessingProcedures of Image Processing
2.2. Image Processing SystemsImage Processing Systems
3.3. Digital Image Processing – CorrectionDigital Image Processing – Correction
4.4. Digital Image Processing – ConversionDigital Image Processing – Conversion
5.5. Digital Image Processing – ClassificationDigital Image Processing – Classification
FCE 552: Engineering Survey IV
Digital Image Processing
Dept. of Geospatial & Space Technology
University of NairobiEngineering Surveying IV
Procedures of Image Procedures of Image ProcessingProcessing
FCE 552: Engineering Survey IV
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Data Flow in Remote Sensing
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Data Processing in Remote Sensing
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Image Processing SystemsImage Processing Systems
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Image Processing Systems with Personal Computer
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Image Processing Systems with Network System
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Digital Image Processing - Digital Image Processing - CorrectionCorrection
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Due to variations in scene illumination and Due to variations in scene illumination and
viewing geometry, atmospheric conditions, viewing geometry, atmospheric conditions,
and sensor noise and response will vary and sensor noise and response will vary
depending on the specific sensor and depending on the specific sensor and
platform used to acquire the data and the platform used to acquire the data and the
conditions during data acquisition; conditions during data acquisition;
It may be desirable to convert and/or It may be desirable to convert and/or
calibrate the data to known (absolute) calibrate the data to known (absolute)
radiation or reflectance units to facilitate radiation or reflectance units to facilitate
comparison between data. comparison between data.
Radiometric Distortion
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Elements of Radiometric Correction
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Atmospheric Effects
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Atmospheric Correction
Dept. of Geospatial & Space Technology
University of Nairobi
Due to several factors, including: the Due to several factors, including: the
perspective of the sensor optics; the motion perspective of the sensor optics; the motion
of the scanning system; the motion of the of the scanning system; the motion of the
platform; the platform altitude, attitude, and platform; the platform altitude, attitude, and
velocity; the terrain relief; and, the velocity; the terrain relief; and, the
curvature and rotation of the Earth. curvature and rotation of the Earth.
To correct for these errors geometric To correct for these errors geometric
registration of the imagery to a known registration of the imagery to a known
ground coordinate system must be ground coordinate system must be
performed i.e., performed i.e., georeferencing.georeferencing.
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Geometric Distortions (1)
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Geometric Distortions (2)
Dept. of Geospatial & Space Technology
University of Nairobi
Flow of Geometric Correction
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Dept. of Geospatial & Space Technology
University of Nairobi
Relationship between Image andGround Control Systems
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Dept. of Geospatial & Space Technology
University of Nairobi
Example of Geometric Correction
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Digital Image Processing - Digital Image Processing - ConversionConversion
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Types of Image Conversion
Dept. of Geospatial & Space Technology
University of Nairobi
Image Enhancement – Grey Scale Conversion
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Dept. of Geospatial & Space Technology
University of Nairobi
Image Enhancement – Histogram Equalization
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Dept. of Geospatial & Space Technology
University of Nairobi
Image Enhancement - Color Composition
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Dept. of Geospatial & Space Technology
University of Nairobi
Feature Extraction - Spatial Filtering
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Dept. of Geospatial & Space Technology
University of Nairobi
Normalized Difference Vegetation Index
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Dept. of Geospatial & Space Technology
University of Nairobi
Principal Component Analysis
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
The objective is to
reduce the dimensionality
in the bands thus
maximizing the amount of
information from the
original data into the
least number of new
components.
Dept. of Geospatial & Space Technology
University of Nairobi
Principal Component Analysis
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Digital Image Processing - Digital Image Processing - ClassificationClassification
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Definition of Classification
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Concept of Classification
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Types of Classifications
Supervised classificationSupervised classification where the operator where the operator
defines the clusters during the defines the clusters during the training process training process
using training data obtained from using training data obtained from ground truthground truth..
Unsupervised classificationUnsupervised classification involves a involves a
clustering algorithmclustering algorithm automatically finding and automatically finding and
defining a number of clusters in the feature defining a number of clusters in the feature
space using only image characteristics or space using only image characteristics or
features.features.
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Features for Classification
Multi-spectral featuresMulti-spectral features
Multi-temporal featuresMulti-temporal features
TextureTexture
Height information (DTM/DEM)Height information (DTM/DEM)
Indices (e.g., NDVI)Indices (e.g., NDVI)
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Procedures of Classification
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Comparison between different classifiers
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Advanced Classifications
Fuzzy ClassificationFuzzy Classification
Contextual ClassificationContextual Classification
Artificial Neural Network ClassificationArtificial Neural Network Classification
Cognition-based ClassificationCognition-based Classification
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Recognition of Man-made Objects
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Example of Road Recognition
Dept. of Geospatial & Space Technology
University of Nairobi
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)
Example of Semi-automated Recognition of Houses
Dept. of Geospatial & Space Technology
University of Nairobi
Depending on the application, the information of Depending on the application, the information of
classes of interest need to be defined and their classes of interest need to be defined and their
spatio-temporal characteristics assessed. On the spatio-temporal characteristics assessed. On the
basis of these, the most appropriate data can be basis of these, the most appropriate data can be
selected and the criteria include:-selected and the criteria include:-
Sensor type (resolution, level of detail, etc).Sensor type (resolution, level of detail, etc).
Relevant wavelength bands.Relevant wavelength bands.
Date of acquisition (important for phenomena that are Date of acquisition (important for phenomena that are
time dependent, also considerations in regard to cloud time dependent, also considerations in regard to cloud
cover and amount of illumination should be taken into cover and amount of illumination should be taken into
account.account.
Budgetary Criteria.Budgetary Criteria.
Remote Sensing Data Selection Criteria
FCE 552: Engineering Survey IV
B. Sc. (Civil Engineering)