Study on urban green space extracting and dynamic monitoring method

Urban green space plays an important and positive role in global carbon cycle, and it is very important for improving the ecological environment of the city, reducing the urban heat island effect, promoting the harmonious relationship between human and nature. In order to overcome the disadvantages of the existing method to extract green space information and solve the problem that consume much time and labor when changing detection of urban green space, and that is difficult to be monitored as well. This paper takes Xi'an urban green space as the researching object, using the LANDSAT TM satellite images, and making use of decision tree classifier based on simple rules and the classification method of support vector machine (SVM) respectively to extract the urban green space information and evaluate its accuracy. Then, the paper proposes to apply the fuzzy C-means method (FCM) into the extraction of green space to solve the problems of the mixed pixel, which exist in green space information extraction of the TM image, for example, the important green information of discrete green belt and street trees that are in the area less than 30 meters, cannot be extracted by the hard classification method. The introduced algorithm can calculate the fuzzy subjected value of the pixel in each classification category, and then automatically classify it as class subjected value, which is good to improve the classification accuracy. The research results show that: (1) compared to the decision tree classification method which is based on the simple rules, the classification accuracy of SVM increased by about 15%. However, the extraction of small area of green space information is still incomplete, such as the information of the green belt and street trees; (2) using the FCM algorithm could do a more refined and accurate classification result into different categories of subjected according to the pixel. Small area of green space information can be extracted first-rate and the classification accuracy is improved well. The proposed algorithm can solve the problem of mixed pixel in the green information extraction; (3) the dynamic monitoring results show that from 1995 to 2010, the green space of Xi'an city is significantly reduced, so the urban ecological construction is an urgent need.