Open AccessAutomated unsupervised change detection technique from RGB color image
Author(s) -
Mohamed Gomaa,
Essam Hamza,
Hassan Elhifnawy
Publication year - 2019
Publication title -
iop conference series. materials science and engineering
Language(s) - English
Resource type - Journals
eISSN - 1757-899X
pISSN - 1757-8981
DOI - 10.1088/1757-899x/610/1/012046
Subject(s) - rgb color model , pixel , change detection , artificial intelligence , computer science , computer vision , sensitivity (control systems) , process (computing) , color image , automation , image resolution , image processing , remote sensing , pattern recognition (psychology) , image (mathematics) , geography , engineering , electronic engineering , operating system , mechanical engineering
Change detection is an important process for many applications such as monitoring the effects of environmental hazards, landslides, rock fall and city development projects. RGB images are commonly used as commercial sources of data for monitoring changes visually because they have powerful descriptive information for different features. Automation of detected changes from two RGB images is a challenge because the two images are usually captured in different environments, as temperature, sun angle, clouds, capturing time …etc. The objective of this research is to introduce an automated technique for detecting changes from RGB image based on color channels. The image pixel is represented as a set of its color channels values, R, G and B which is called color signature of image pixel. A real data is used to fulfil the research objective without pre-knowledge about the changes. The correlation coefficients are calculated between color signatures of each two associated pixels from two different registered high resolution satellite images for the same area of study. The detected pixels of changes are identified based on specific correlation value that is identified based on degree of change sensitivity. The degree of sensitivity is based on the importance of detection procedure that is considered as a main part of decision making for risk and crisis management system. The proposed technique is unsupervised and fully automated. It can be applied through a real time process based on the processing capabilities, size and resolution of input images. This technique is easy to use and gives accurate results with neglecting the effects of atmospheric effects.