PROGRESS IN GEOGRAPHY ›› 2016, Vol. 35 ›› Issue (1): 89-97.doi: 10.18306/dlkxjz.2016.01.010

• Orginal Article • Previous Articles     Next Articles

Case-based formalization and inference method of application-matching knowledge on digital terrain analysis

Xuewei WU1,2(), Chengzhi QIN1,3,*(), Axing ZHU3,4   

  1. 1. State Key Laboratory of Resources and Environmental Information System, Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing 100101, China
    2. University of Chinese Academy of Sciences, Beijing 100049, China
    3. Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, School of Geography, Nanjing Normal University, Nanjing 210023, China
    4. Department of Geography, University of Wisconsin-Madison, Madison, Wisconsin 53706, USA
  • Online:2016-01-31 Published:2016-01-31
  • Contact: Chengzhi QIN;
  • Supported by:
    National Natural Science Foundation of China, No.41422109, No.41431177


Application of digital terrain analysis (DTA) relies heavily on the DTA-domain knowledge on the match between the chosen algorithm (and its parameter-settings) and the application context (such as target task, terrain condition of the study area, and DEM resolution)—the so-called application-matching knowledge. This type of knowledge has a direct impact on the quality of DTA modelling when users of DTA, especially non-expert users, do not have sufficient amount of such knowledge to support their DTA applications. Existing DTA-assisted tools often cannot use application-matching knowledge because this type of knowledge has not been formalized in DTA to be available for inference in these tools. This is mainly because this type of DTA knowledge is currently inaccurate and non-systematic, and often exists in documents for specific case studies, or as personal knowledge of domain experts. This situation makes the DTA modelling process difficult for users, especially for non-expert users. Case-based reasoning method that originated from artificial intelligence is appropriate for formalization and inference of non-systematic knowledge. In this article, we propose a case-based formalization and inference method for the application-matching knowledge in DTA. The specific design of the proposed case-based method can be divided into two parts: formalization of the application-matching knowledge, and inference method. The case of this knowledge consists of a series of indices to formalize the DTA application-matching knowledge and the corresponding similarity calculation methods for inference based on the case. To evaluate the performance of the proposed method, we implemented it in a software prototype of DTA modelling environment and then applied it to a DTA application of river network extraction. In the experiment we prepared 32 cases of river network extraction in the USA. The results of cross validation preliminarily show that the proposed case-based method is suitable for using the application-matching knowledge in DTA. It reduced the modelling burden greatly for users.

Key words: Digital Terrain Analysis (DTA), modelling knowledge, case, river network extraction