An Efficient Approach for Sky Detection Irfanullah 1 , Kamal Haider 2 , Qasim Sattar 1 , Sadaqat-ur-Rehman 1 , Amjad Ali 1 1 Sarhad University of Science & Information Technology (SUIT), Peshawar, Pakistan 2 Gandhara Institute of Science and Technology, Peshawar Abstract— Blue color has been proven to be a useful and robust cue for sky detection, localization and for tracking RGB color in different applications of image processing. In this paper a pixel based solution utilizing the sky color information has been proposed for sky detection. The sky color information is extracted through the comparison of RGB values of a pixel. Based on the experimental results on highly complex still images, our approach for sky detection has been proved to be accurate, fast and simple. Key words: Sky detection, RGB, pixel based sky detection, Color spaces. 1. INTRODUCTION The sky detection is not a new problem for the researchers in the image processing domain. Due to its vast range of applications in weather forecasting, solar exposure prediction, image acquisition and understanding and in image retrieval and orientation, sky detection has been a keen area for researchers. Sky detection becomes very difficult under certain circumstances, especially in overcast conditions and different types of clouds makes a real challenge for sky detection algorithms. Mostly the pixel based approaches are used for sky detection in images/ videos to improve the image/ video quality as it becomes very easy to predict noise in the sky regions in an image because of its smooth appearance. The color is considered the most robust and accurate feature for the sky detection. In this paper, we proposed a three step pixel based approach to sky detection that incorporates the RGB values for pixels classification. Simple if else conditions are used for the identification of sky region pixels in an image. 2. RELATED WORK Due to various applications like video/image quality enhancement, solar exposure prediction and weather forecasting, sky detection has been thoroughly studied and tackled through various approaches. Zafarifar et al. [1] represent a novel approach for sky detection using two different features. This algorithm utilizes the adaptive positioning and color modeling for segmentation and extracting the sky region information in an image/video. The proposed algorithm produced better performance compared to state-of-the-art approaches in sky detection in natural scenes. Schmitt et al. [2] used color, position and shape as features for the sky detection in their approach. The performance of the proposed algorithm was tested on a number of outdoor images and the based on the analysis of different of the experimental results under different weather and lighting conditions the author claim for highly accurate performance in classifying the sky regions in images taken in clear, overcast and partially clouded weather. Laungrungthip et al. [3] represent a solar exposure system based on the image processing techniques. Image processing algorithms are used for segmentation of the outdoor images taken under different lighting conditions to segment out the sky regions in a scene. A robust approach composed of canny edge detection algorithm [4] and Morphological closing algorithm [5], was adopted for identifying and separating the sky regions in color images. IJCSI International Journal of Computer Science Issues, Vol. 10, Issue 4, No 1, July 2013 ISSN (Print): 1694-0814 | ISSN (Online): 1694-0784 www.IJCSI.org 222 Copyright (c) 2013 International Journal of Computer Science Issues. All Rights Reserved.