|
|
Detection of biological structures in TEM microscope images
Cikánek, Martin ; Chmelík, Jiří (referee) ; Potočňák, Tomáš (advisor)
The aim of the first part of this thesis is to explain the theoretical basis of transmission electron microscopy and to mention fundamental parts of transmission electron microscopes. The next part of this work is focused on possible methods of image segmentation, the use of neural networks in the detection of objects in an image and the subsequent clustering of results. The theoretical part of the thesis is concluded with an explanation of some already published methods of automatic detection of biological structures in microscopic images and theoretical design of the algorithm, which will be subsequently developed. The process of training neural networks in order to automatically detect biological structures in an image is described at the beginning of the practical part. This is followed by an evaluation of the results achieved by these networks. Subsequently, cluster analysis methods are applied to these results, the products of which are compared with each other and also with the results obtained by already published methods.
|
|
|
Unsupervised learning
Kantor, Jan ; Sáblík, Václav (referee) ; Honzík, Petr (advisor)
The purpose of this work has been to describe some techniques which are normally used for cluster data analysis process of unsupervised learning. The thesis consists of two parts. The first part of thesis has been focused on some algorithms theory describing advantages and disadvantages of each discussed method and validation of clusters quality. There are many ways how to estimate and compute clustering quality based on internal and external knowledge which is mentioned in this part. A good technique of clustering quality validation is one of the most important parts in cluster analysis. The second part of thesis deals with implementation of different clustering techniques and programs on real datasets and their comparison with true dataset partitioning and published related work.
|
|
|
Detection of biological structures in TEM microscope images
Cikánek, Martin ; Chmelík, Jiří (referee) ; Potočňák, Tomáš (advisor)
The aim of the first part of this thesis is to explain the theoretical basis of transmission electron microscopy and to mention fundamental parts of transmission electron microscopes. The next part of this work is focused on possible methods of image segmentation, the use of neural networks in the detection of objects in an image and the subsequent clustering of results. The theoretical part of the thesis is concluded with an explanation of some already published methods of automatic detection of biological structures in microscopic images and theoretical design of the algorithm, which will be subsequently developed. The process of training neural networks in order to automatically detect biological structures in an image is described at the beginning of the practical part. This is followed by an evaluation of the results achieved by these networks. Subsequently, cluster analysis methods are applied to these results, the products of which are compared with each other and also with the results obtained by already published methods.
|
|
|
Clustering objects with the MCluster-Miner procedure of the LISp-Miner system
Pelc, Tomáš ; Šimůnek, Milan (advisor) ; Šulc, Zdeněk (referee)
This bachelor thesis deals with clustering objects with the MCluster-Miner procedure of the LISp-Miner system. The first aim of this bachelor thesis is clustering objects with the mentioned pro-cedure and analyzing its possible usage on different datasets. To achieve this goal, the procedure was applied on six different datasets. The secong aim of this thesis is to analyze and compare implemented algorithms, similarity measures and to propose recommendations for clustering parameters. To achieve this goal, the available algorithms and similarity measures are compared based on achieved results (the quality of distribution objects into clusters, the time of clustering task, the number of attributes used for clustering). Based on these comparisons, the recommen-dations for clustering parameters are proposed. The benefits of this thesis are these recommenda-tions, comparisons of available algorithms and similarity measures, summary of actual state (da-ted to May 2017) of the MCluster-Miner module and showing the possibility of displaying results of clustering task at the interactive analysis of geodata. The theoretical part comprises the description of the LISp-Miner system, basic clustering principles, clustering methods and similari-ty measures used by the GUHA-procedure MCluster-Miner, and the MCluster-Miner module. In the practical part the MCluster-Miner procedure is being applied on six different datasets and the achieved results are summarized there.
|
|
|
Unsupervised learning
Kantor, Jan ; Sáblík, Václav (referee) ; Honzík, Petr (advisor)
The purpose of this work has been to describe some techniques which are normally used for cluster data analysis process of unsupervised learning. The thesis consists of two parts. The first part of thesis has been focused on some algorithms theory describing advantages and disadvantages of each discussed method and validation of clusters quality. There are many ways how to estimate and compute clustering quality based on internal and external knowledge which is mentioned in this part. A good technique of clustering quality validation is one of the most important parts in cluster analysis. The second part of thesis deals with implementation of different clustering techniques and programs on real datasets and their comparison with true dataset partitioning and published related work.
|
|
|
Cluster analysis as a tool for object classification
Vanišová, Adéla ; Löster, Tomáš (advisor) ; Bílková, Diana (referee)
The aim of this thesis is to examine the cluster analysis ability segment the data set by selected methods. The data sets are consisting of quantitative variables. The basic criterion for the data sets is that the number of classes has to be known and the next criterion is that the membership of all object to each class has to be known too. Execution of the cluster analysis was based on knowledge about the number of classes. Classified objects to individual clusters were compared with its original classes. The output was the relative success of classification by selected methods. Cluster analysis methods are not able to determine an optimal number of clusters. Estimates of the optimal number of clusters were the second step in analysis for each data set. The ability of selected criteria identify the original number of classes was analyzed by comparing numbers of original classes and numbers of optimal clusters. The main contribution of this thesis is the validation of the ability of selected cluster analysis methods to identify similar objects and verify the ability of selected criteria to estimate the number of clusters corresponding to the real file distribution. Moreover, this work provides a structured overview of the basic cluster analysis methods and indicators for estimating the optimal number of clusters.
|
|
|
Evaluation of Cluster Analysis Methods
Löster, Tomáš ; Řezanková, Hana (advisor) ; Berka, Petr (referee) ; Dohnal, Gejza (referee)
Cluster analysis includes a range of methods and practices that are used primarily for classification of objects. It takes an important role in many areas. Since the resulting distribution of objects into clusters may vary depending on the selected methods and specifications, it is appropriate to assess the results obtained. This paper proposes new ways of evaluating these results in a situation where objects are characterized by qualitative variables or by variables of different types. These coefficients can be used either to compare different methods (in terms of better outcomes) or for finding of the optimal number of clusters. All of them are based on the detection of variability which is also used for measuring of dissimilarity of objects and clusters. The newly proposed evaluation methods are applied to real data sets (of different sizes, with different number of variables, including variables of different types) and the behavior of these coefficients in different conditions is being examined. These data sets have known as well as unknown classification of objects into clusters. The best coefficient for evaluating clustering results with different types of variables can be considered, based on the analysis carried out, the modified coefficient of CHF. Local maximum value according to which the results of the clustering are evaluated, almost always exists. The analysis has proven that in most cases this value meets the expected results of the well-known classification of objects into clusters. The existence of local extremes of the other coefficients depends on specific data sets and is not always feasible.
|
guest ::
