Kun Qin

Sea surface temperature clustering based on type-2 fuzzy theory

Spatial data clustering is an effective method to find interesting spatio-temporal clustering patterns. There are many uncertainties in sea surface temperature (SST) clustering, so clustering methods with uncertainy must be used. Type-2 fuzzy theory takes into account the uncertainty of membership grade while fuzzy C means (FCM) not. Based on the analysis of interval type-2 fuzzy C means (IT2FCM), the paper utilizes two normal cloud models to express fuzzifiers m1 and m2, and uses two cloud drops to substitute them. The method considers the uncertainty of two fuzzifiers, and avoids many times of repeated tests, which reduces computation cost. The paper applies the improved IT2FCM into global SST clustering, and discovers some interesting climate patterns.

On the methods of image segmentation with uncertainty

There are much uncertainty in the process of image segmentation, The paper firstly researched the sources of uncertainty of image segmentation; and then analyzed the method of image segmentation based on K means cluster, and the method based on fuzzy K means cluster; and then, the paper researched the theory of cloud model, which considers the fuzziness, random and the their association of uncertainty. The paper put forward a new method of image segmentation based on cloud model. Lastly, the experiments proved the method of image segmentation based on cloud model is better than the method based on fuzzy K means cluster and the method based on K means.

Built-up area extraction using data field from high-resolution satellite images

Built-up areas are typical man-made structure in urban environment, and timely and accurately acquiring built-up area layers can provide necessary geo-spatial information for planners and policymakers. In this paper, a data field-based method is proposed for the automated detection of built-up areas from high-resolution satellite images. This method views the local corner features of buildings as mass points and models their spatial interaction and distribution by potential function. Due to the salient potential differences between built-up areas and non-built-up areas, the built-up areas are extracted by threshold based segmentation. Two further post-processing techniques, that is, noise removal and hole filling, significantly improve the detection results. The experimental results indicate that the proposed method shows very good performance, and it, with simple parameter settings and without needing sample information, not only can achieve high extraction accuracy but also can effectively keep the shape and topological structure of built-up areas by further post-processing techniques.

Cloud model based fuzzy C-means clustering and its application

The Algorithm of Fuzzy C-Means (FCM) clustering is used in many fields, such as data mining, image segmentation etc. But it has the problem of cluster center initialization. Good initial cluster centers will constrain the value function to the overall situation optimal solution rapidly, and inappropriate initial cluster centers, not only need more iterative times, but also may possibly cause the algorithm finally restrained to the partial optimal solution. Aim to resolve the problem of cluster center initialization, the paper proposes a new approach of FCM based on cloud model which is an efficient transformation model between quantitative number and qualitative concept, and applied it in the field of image segmentation, the experiment results prove the method can define good initial cluster centers and produce good quality of image segmentation.

Spatio-temporal data clustering based on type-2 fuzzy sets and cloud models

The time series remote sensing data and meteorological satellite data offer new opportunities for understanding the earth system. Spatio-temporal data clustering becomes a kind of idea tool to explore huge data space of spatio-temporal data. Because there are many uncertainties in the huge spatio-temporal data, including fuzziness and randomness, the spatio-temporal clustering methods with uncertainties are needed. Based on type-2 fuzzy sets and cloud models, the paper analyzes the uncertainty of the membership of FCM (fuzzy C-means), and proposes CFFCM (cloud fuzzifier fuzzy C-means) method. Take the time series SST (sea surface temperature) data as examples, the paper applies CFFCM to carry out spatio-temporal clustering analysis, and discovers some interesting patterns.