An Image Processing Technique for the Study of Urban Heat Island Changes Using Different Seasonal Remote Sensing Data

Thermal remote sensing imagery has been widely used to reveal urban heat island (UHI) phenomenon since the last two decades. To study the UHI change through change detection technique, it is usually expected to use the thermal data of different years with the same date. However, this is generally difficult special for the cloudy and rainy areas and the data with different date may have to be used. This will make the comparison of such thermal image data more difficult due to seasonal difference. In order to work out this problem, this paper introduces a method which can reduce the seasonal difference if the images acquired on different date have to be used. \;The thermal infrared bands of different date were first processed respectively through several image enhancement technologies. This generated 3\|dimension\|view images and revealed heat characteristics and spatial distribution features of the UHI. To find out the change of the UHI between different dates, the two thermal images were normalized and scaled to several grades to reduce seasonal difference and then overlaid to produce a difference image by subtracting corresponding pixels. To quantitatively compare UHI, an index, Urban\|Heat\|Island Ratio Index (URI), was created. It can further reveal the intensity of the UHI within the urban area. The calculation of the index is based on the ratio of UHI area to urban area. The greater the index is, the more intense the UHI is. This makes it possible to quantitatively study the UHI changes within a period of time. The technique has been successfully applied in a case study on UHI change in Xiamen City of China.