Date of Award


Degree Name

Master of Science


Electrical Engineering


Sergio D. Cabrera


In this research, we propose spatio-spectral processing techniques for the detection of dust storms in 5-band NOAA-AVHRR imagery. The research is mainly focused on locating the dust sources and automatically finding the transport direction of the dust storm. Previous methods that use simple band math analysis have produced promising results but have some drawback in producing consistent results when low SNR images are used. Moreover, in seeking to automate the dust storm detection, the presence of clouds in the vicinity of the dust storm creates a challenge in being able to distinguish these two types of image texture. Our research not only seeks to detect the presence of the dust storm in the imagery, it also attempts to find the directionality and to locate the sources of the dust storm. We propose a spatio-spectral processing scheme with two techniques for this detection scheme - visualization and automation. Visualization technique is intended to locate the dust sources and automation technique is proposed to detect the direction of the dust storm.

For visualization technique, image processing algorithms like the spectral-domain PCA/MNF transforms, and the spatial-domain k-means unsupervised classification method, are evaluated as tools to help us find dust storms by visual image analysis. For locating the dust sources based on image information, directional filtering is used in combination with edge detectors and spectral-domain classification technique.

Next, edge detectors like Sobel and Frei-Chen are applied to the selected filtered images for further enhancement of the streaks produced by the directional texture. False color composite images are created to visually enhance the directional streaks to be able to locate the dust sources. The automation technique for finding the direction of the dust storm involves performing the power spectrum analysis on bands 4 and 5 since these wavelengths highlight the absorption and subsequent emission of thermal radiation by the silicate particles in the dust storms. The processing scheme involves block processing applied for power spectrum analysis followed by binary thresholding and morphological enhancement. A local power spectral density analysis is first used to confirm the presence of high directionality information in certain regions of an image. These regions can be determined to be candidate dust storm regions. Binary thresholding is performed on these blocks to enhance the directional texture. Morphological enhancement is done on these binary images to compute their area and orientation.




Received from ProQuest

File Size

106 pages

File Format


Rights Holder

Swapna Janugani