DefinePK

Abstract

This paper presents a development of an automatic system that's able to detect indoor air thermal pollution automatically by using thermal image processing technique. This research was proposed to enable civilians to improve the quality of indoor air to ensure the building is well-ventilated and getting rid of air pollutants. Indoor air is more dangerous than outdoor air as civilians spend most of the time in indoor. 255 images are used in this simulation process whereby the images are divided into two categories. The first category has been classified consists of 178 images with the colour changes (pollution detected) while another 77 images showed with no colour changes (no pollution detected). Colour conversion, image enhancement and morphological operation were used in this research in order to process the desired output image. The accuracy of the system was measured for the detection of wet or moist wall that will affect the indoor air quality on polluted data shows 94.94%. While 100% accuracy was achieved for no wet or moist walls detected. The experiment results indicate that this proposed method able to detect indoor air thermal pollution on the differences of moisture area and the wet areas.