Scribd
Upload
53 views

Rough-Set-Based Color Channel Selection

This document proposes using rough sets to select color channels for accurate sky and cloud segmentation in images from ground-based sky cameras. Most prior work used thresholding on arbitrarily selected color channels like red and blue. The proposed method assesses which color channels contribute most to segmentation using rough sets. It defines a decision table where each row represents an observation with attributes like color channels. The experimental results show rough sets identify favorable channels better than other feature selection algorithms.

Uploaded by

Anonymous 1aqlkZ
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
53 views

Rough-Set-Based Color Channel Selection

This document proposes using rough sets to select color channels for accurate sky and cloud segmentation in images from ground-based sky cameras. Most prior work used thresholding on arbitrarily selected color channels like red and blue. The proposed method assesses which color channels contribute most to segmentation using rough sets. It defines a decision table where each row represents an observation with attributes like color channels. The experimental results show rough sets identify favorable channels better than other feature selection algorithms.

Uploaded by

Anonymous 1aqlkZ
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 4

Rough-Set-Based Color Channel Selection

Abstract

Color channel selection is essential for accurate segmentation of sky and clouds in images
obtained from ground-based sky cameras. Most prior works in cloud segmentation use threshold-
based methods on color channels selected in an ad hoc manner. In this letter, we propose the use
of rough sets for color channel selection in visible-light images. Our proposed approach assesses
color channels with respect to their contribution for segmentation and identifies the most
effective ones.

Index Terms— Cloud analysis, color channel, rough set, whole sky imagers (WSIs).

Objective

In this method to separate the sky and cloud color channel segmentation is done. Rough-sets are
proposed to segment the sky and color.

1. INTRODUCTION:

Ground-based whole sky imagers (WSIs) are becoming popular among the remote sensing
community. They provide instantaneous data of cloud formations and are thus useful in a variety
of applications . WSIs complement satellite images with localized data of higher temporal and
spatial resolution. They capture images of the sky at regular intervals and archive them for
processing. Most WSIs use traditional cameras in the visible-light spectrum, while a few models
capture the near-infrared range as well .

Typical post processing algorithms include the computation of the fraction of the sky covered by
clouds, the recognition of cloud types, or the estimation of the cloud base height. A prerequisite
for these applications is the segmentation of clouds from the sky, with each pixel of the image
classified as either sky or cloud. As the sky is predominantly blue because of Rayleigh
scattering, most existing approaches use thresholding on an ad hoc combination of red and blue
color channels. Kreuter et al. used a fixed threshold for the blue/red ratio. Calbó and Sabburg use
various statistical features (mean, standard deviation, entropy, etc.) obtained from red and blue
channels for successful detection and subsequent labeling of pixels. The difference of red and
blue channels is exploited. The saturation channel of the HSV color model is used. Recently, Li
et al. proposed the use of a normalized red/blue ratio for cloud detection.

Proposed method

Fig. 1. Illustration of a typical rough set, which approximates a conventional set X (depicted in
blue). Each individual grid depicts a partition of the universe generated by an equivalence
relation. The union of all such partitions indicated with green solid borders (definite members)
represents the lower approximation P(X), while the red dotted borders (possible members)
represent the upper approximation P(X).

In rough sets, information is expressed in the form of a decision table. We define a decision table
L such that each row represents an observation and each column is an attribute from attribute set
A. This nonempty set of observations is usually referred to as the universe U.

4. SOFTWARE AND HARDWARE REQUIREMENTS

 Operating system : Windows XP/7.


 Coding Language : MATLAB
 Tool : MATLAB R 2012

SYSTEM REQUIREMENTS:
HARDWARE REQUIREMENTS:

 System : Pentium IV 2.4 GHz.


 Hard Disk : 40 GB.
 Floppy Drive : 1.44 Mb.
 Monitor : 15 VGA Colour.
 Mouse : Logitech.
 Ram : 512 Mb.

5. CONCLUSION:

Sensing the earth’s atmosphere using ground-based visible-light images is popular because of its
low cost and the high temporal and spatial resolution of the captured images, compared with
traditional satellite images. In this letter, we have proposed a color channel selection algorithm
based on rough set theory. The experimental results show the efficacy of our approach in
identifying favorable color channels for image segmentation. Our proposed approach
outperforms other feature selection algorithms. Future work involves the extension of such
rough-set-based approaches to other applications.

6. REFERENCES:

[1] S. Dev, B. Wen, Y. H. Lee, and S. Winkler, “Ground-based image analysis: A tutorial on
machine-learning techniques and applications,” IEEE Geosci. Remote Sens. Mag., vol. 4, no. 2,
pp. 79–93, Jun. 2016.

[2] S. Dev, F. M. Savoy, Y. H. Lee, and S. Winkler, “Design of low-cost, compact and weather-
proof whole sky imagers for high-dynamic-range captures,” in Proc. Int. Geosci. Remote Sens.
Symp. (IGARSS), Jul. 2015, pp. 5359–5362.

[3] E. Rumi, D. Kerr, J. M. Coupland, A. P. Sandford, and M. J. Brettle, “Automated cloud


classification using a ground based infra-red camera and texture analysis techniques,” Proc.
SPIE, vol. 8890, Oct. 2013, Art. no. 88900J.
[4] S. Dev, F. M. Savoy, Y. H. Lee, and S. Winkler, “WAHRSIS: A low-cost high-resolution
whole sky imager with near-infrared capabilities,” Proc. SPIE, vol. 9071, May 2014, Art. no.
90711L.

[5] Q. Li, W. Lu, J. Yang, and J. Z. Wang, “Thin cloud detection of all-sky images using Markov
random fields,” IEEE Geosci. Remote Sens. Lett., vol. 9, no. 3, pp. 417–421, May 2012.

You might also like