No abstract

No abstract

No abstract

In this paper, we present a parallel computing technique, referred to as parallel positive Boolean function (PPBF), for supervised classification of multispectral images. The approach is based on the generalized positive Boolean function (GPBF) scheme, which has been successfully applied in multispectral image classification. The GPBF classifier is developed from a stack filter. The stack filter is defined as the class of all nonlinear digital filters. Each stack filter corresponding to a GPBF possesses the weak superposition property and the ordering property. In order for the GPBF to be effective, the proposed PPBF is performed to improve the computational speed by using parallel cluster computing techniques. It creates a set of stack filters in each parallel node implemented by message passing interface (MPI). The proposed PPBF technique reduces the structure complexity of original GPBF. The effectiveness of the proposed PPBF is evaluated by fusing Systeme Pour l'Observation de la Terre (SPOT) images and digital elevation model (DEM) information for land cover classification during the post 921 Earthquake period inTaiwan. The experimental results demonstrated that PPBF not only significantly improves the computational loads of GPBF classification, but also substantially improves the precision of classification compared to conventional classification.

Taiwan Island is of mountainous with high density of population concentrated on a narrow belt of western plain. Human activities are pushing to move toward hillside and even mountains after an overdeveloping of flat plain. These include local community, agricultural zone, golf course, and road network, among others. The competition of land use with nature leads to landscape change to dramatic degree. The complex geological setting of is prone to landslide and soil erosion triggered by torrent storms if soil and water conversation are not well cared. The 921 Great Earthquake further made the soil even more vulnerable to slide and collapse. This has been seen from floods in 2004 which caused large scale damages in middle Taiwan. Hence, monitoring of hillside change becomes critical for effective and efficient land management. These changes are of spatial and temporal variability. In this paper, we analyze the statistical properties of land cover changes detecting by SPOT images from two years of continuous observations and monitoring, with ancillary data from base maps, land use maps, GIS data, and DTM. All the data handling and analysis are through a GIS system. To assess the detection accuracy, a total of well distributed 206 samples from the whole island are selected for verification. Those changed areas are then analyzed in terms of occurrence frequency associated with location and time. Spatially, the occurrence frequency is high for slope between 15%-30% with altitude of around 1500m. The occurrence frequency was also dependent on the geological risk sites and strongly related to landslide-prone sites. Preliminary, it was found that the correlation between space and time is weak. Longer observation in time may be necessary and is undergoing.