Decision tree classification of land cover from remotely sensed data

Decision tree classification algorithms have significant potential for land cover mapping problems and have not been tested in detail by the remote sensing community relative to more conventional pattern recognition techniques such as maximum likelihood classification. In this paper, we present several types of decision tree classification algorithms arid evaluate them on three different remote sensing data sets. The decision tree classification algorithms tested include an univariate decision tree, a multivariate decision tree, and a hybrid decision tree capable of including several different types of classification algorithms within a single decision tree structure. Classification accuracies produced by each of these decision tree algorithms are compared with both maximum likelihood and linear discriminant function classifiers. Results from this analysis show that the decision tree algorithms consistently outperform the maximum likelihood and linear discriminant function classifiers in regard to classf — cation accuracy. In particular, the hybrid tree consistently produced the highest classification accuracies for the data sets tested. More generally, the results from this work show that decision trees have several advantages for remote sensing applications by virtue of their relatively simple, explicit, and intuitive classification structure. Further, decision tree algorithms are strictly nonparametric and, therefore, make no assumptions regarding the distribution of input data, and are flexible and robust with respect to nonlinear and noisy relations among input features and class labels.

Research on extracting method of micro-scale remote sensing information combination and application in coastal zone

973 Project of China [2006CB701305]; "863" Project of China [2009AA12Z148]; National Natural Science Foundation of China [40971224]

SATELLITE REMOTE-SENSING PREDICTION OF JAPANESE PILCHARD FISHING GROUND IN THE HUANGHAI SEA AND THE EAST-CHINA-SEA

The thermophily, fishing season and central fishing ground of Japanese pilchard (Sardinops melanosticta) were studied by using satellite remote sensing (SRS) and other methods in Haizhou Bay and Tsushima waters during 1986-1990. A rapid prediction method of fishing ground is presented. Moreover, the results indicated that the thermophilic values of the fish stock are 11-20 degrees C and both fishing grounds are in increasing temperature process from the beginning to the end of the fishing period. The Japanese pilchards gather vigorously at the sea surface temperature of 15-17 degrees C. The water temperature is a key factor affecting the fishing season and the catch of the fishing ground. The increasing temperature process restricts the fishing season development and central fishing ground formation. The accuracy of 15 predictions made in the Haizhou Bay fishing ground is up to 91.3%, and 37 predictions made in the Tsushima, fishing ground shorten the fish detection time by 13.4% - 22% on the average.

Opportunities for X-ray remote sensing at Mercury

Grande, Manuel; Dunkin, S. K.; Kellett, B., 'Opportunities for X-ray remote sensing at Mercury', Planetary And Space Science (2001) 49(14-15) pp.1553-1559 RAE2008

Hazard and risk assessment from Hurricane Ivan (2004) in Grenada using Geographical Information Systems and remote sensing

Hurricanes are destructive storms with strong winds, intense storm surges, and heavy rainfall. The resulting impact from a hurricane can include structural damage to buildings and infrastructure, flooding, and ultimately loss of human life. This paper seeks to identify the impact of Hurricane Ivan on the aected population of Grenada, one of the Caribbean islands. Hurricane Ivan made landfall on 7th September 2004 and resulted in 80% of the population being adversely aected. The methods that were used to model these impacts involved performing hazard and risk assessments using GIS and remote sensing techniques. Spatial analyses were used to create a hazard and a risk map. Hazards were identied initially as those caused by storm surges, severe winds speeds, and flooding events related to Hurricane Ivan. These estimated hazards were then used to create a risk map. An innovative approach was adopted, including the use of hillshading to assess the damage caused by high wind speeds. This paper explains in detail the methodology used and the results produced.