Ary Noviyanto

Evaluation of data integration strategies based on kernel method of clinical and microarray data.

The cancer classification problem is one of the most challenging problems in bioinformatics. The data provided by Netherland Cancer Institute consists of 295 breast cancer patient; 101 patients are with distant metastases and 194 patients are without distant metastases. Combination of features sets based on kernel method to classify the patient who are with or without distant metastases will be investigated. The single data set will be compared with three data integration strategies and also weighted data integration strategies based on kernel method. Least Square Support Vector Machine (LS-SVM) is chosen as the classifier because it can handle very high dimensional features, for instance, microarray data. The experiment result shows that the performance of weighted late integration and the using of only microarray data are almost similar. The data integration strategy is not always better than using single data set in this case. The performance of classification absolutely depends on the features that are used to represent the object.

Evaluation of SIFT and SURF features in the songket recognition

The songket recognition is a challenging task. The SIFT and SURF, which are feature descriptors, are considered as potential features for pattern matching. The Songket is a special pattern originally from Indonesia; The Songket Palembang is used in this research. One motif in the Songket Palembang may has several different basic patterns. The matching scores, i.e., distance measure and number of keypoint, are evaluated corresponding with the SIFT and SURF method. SIFT method has been better than SURF method, but SURF has been extremely faster than SIFT.

Cattle's fur detection in complex background based on Graph Cuts

Segmentation becomes a difficult task if the objects are not homogeneous and have overlapping characteristics. The Graph Cuts methods combined with Gaussian Mixture Model (GMM) for initialization label has been adopted to detect cattle object in an image with complex background. The RGB colors and Gray Level Co-occurrence Matrix (GLCM) textures are used as the features set. This method can robustly segment the cattle beef image from its background. This segmentation method produces the average of accuracy value up to 90%.