|
|
Texture Spectral Similarity Criteria Comparison
Havlíček, Michal ; Haindl, Michal
Criteria capable of texture spectral similarity evaluation are presented and compared. From the fifteen evaluated criteria, only four criteria guarantee zero or minimal spectral ranking errors. Such criteria can support texture modeling algorithms by comparing the modeled texture with corresponding synthetic simulations. Another possible application is the development of texture retrieval, classification, or texture acquisition system. These criteria thoroughly test monotonicity and mutual correlation on specifically designed extensive monotonously degrading experiments.
|
|
|
Image Segmentation
Mikeš, Stanislav ; Haindl, Michal (advisor) ; Scarpa, Giuseppe (referee) ; Jan, Jiří (referee)
Image segmentation is a fundamental part in low level computer vision processing. It has an essential influence on the subsequent higher level visual scene interpretation for a wide range of applications. Unsupervised image segmentation is an ill-defined problem and thus cannot be optimally solved in general. Several novel unsupervised multispectral image segmentation methods based on the underlaying random field texture models (GMRF, 2D/3D CAR) were developed. These segmenters use efficient data representations that allow an analytical solutions and thus the segmentation algorithm is much faster in comparison to methods based on MCMC. All segmenters were extensively compared with the alternative state- of-the-art segmenters with very good results. The MW3AR segmenter scored as one of the best available. The cluster validation problem was solved by a modified EM algorithm. Two multiple resolution segmenters were designed as a combination of a set of single segmenters. To tackle a realistic variable lighting in images, the illumination invariant features were derived and the illumination invariant segmenter was developed. For the proper evaluation of segmentation results and ranking of algorithms, a unique web-based texture segmentation benchmark was proposed and implemented. It was used for comprehensive...
|
|
|
Query by Pictorial Example
Vácha, Pavel ; Haindl, Michal (advisor) ; Drbohlav, Ondřej (referee) ; Geusebroek, Jan-Mark (referee)
Ongoing expansion of digital images requires new methods for sorting, browsing, and sear- ching through huge image databases. This is a domain of Content-Based Image Retrieval (CBIR) systems, which are database search engines for images. A user typically submit a query image or series of images and the CBIR system tries to find and to retrieve the most similar images from the database. Optimally, the retrieved images should not be sensitive to circumstances during their acquisition. Unfortunately, the appearance of natural objects and materials is highly illumination and viewpoint dependent. This work focuses on representation and retrieval of homogeneous images, called textu- res, under the circumstances with variable illumination and texture rotation. We propose a novel illumination invariant textural features based on Markovian modelling of spatial tex- ture relations. The texture is modelled by Causal Autoregressive Random field (CAR) or Gaussian Markov Random Field (GMRF) models, which allow a very efficient estimation of its parameters, without the demanding Monte Carlo minimisation. Subsequently, the estimated model parameters are transformed into the new illumination invariants, which represent the texture. We derived that our textural representation is invariant to changes of illumination intensity and...
|
|
|
Texture modeling applied to medical images
Remeš, Václav ; Haindl, Michal (advisor)
and contributions This thesis presents novel descriptive multidimensional Markovian textural models applied to computer aided diagnosis in the field of X-ray mammogra- phy. These general mathematical models, applicable in wide areas of texture modeling outside X-ray mammography as well, provide ideal visual verification using synthesis of the corresponding measured data spaces, contrary to stan- dard discriminative models. All achieved results in the thesis are extensively benchmarked. The thesis presents two methods for breast density classification in X-ray mammography. The methods were tested on the widely known MIAS database and the state-of-the art INbreast database, with competitive results. Several methods for completely automatic mammogram texture enhance- ment are presented. These methods are based on the descriptive textural mod- els developed in the thesis which automatically adapt to the analyzed X-ray texture, thus being universal for any type of input without the need of further manual tuning of specific parameters. The methods' outputs highlight regions of interest, detected as textural abnormalities. The methods provide the pos- sibility of enhancement tuned to specific types of mammogram tissue. Hence, the enhanced mammograms can help radiologists to decrease their false negative...
|
|
|
Query by Pictorial Example
Vácha, Pavel ; Haindl, Michal (advisor)
Ongoing expansion of digital images requires new methods for sorting, browsing, and sear- ching through huge image databases. This is a domain of Content-Based Image Retrieval (CBIR) systems, which are database search engines for images. A user typically submit a query image or series of images and the CBIR system tries to find and to retrieve the most similar images from the database. Optimally, the retrieved images should not be sensitive to circumstances during their acquisition. Unfortunately, the appearance of natural objects and materials is highly illumination and viewpoint dependent. This work focuses on representation and retrieval of homogeneous images, called textu- res, under the circumstances with variable illumination and texture rotation. We propose a novel illumination invariant textural features based on Markovian modelling of spatial tex- ture relations. The texture is modelled by Causal Autoregressive Random field (CAR) or Gaussian Markov Random Field (GMRF) models, which allow a very efficient estimation of its parameters, without the demanding Monte Carlo minimisation. Subsequently, the estimated model parameters are transformed into the new illumination invariants, which represent the texture. We derived that our textural representation is invariant to changes of illumination intensity and...
|
|
|
Texture modeling applied to medical images
Remeš, Václav ; Haindl, Michal (advisor)
and contributions This thesis presents novel descriptive multidimensional Markovian textural models applied to computer aided diagnosis in the field of X-ray mammogra- phy. These general mathematical models, applicable in wide areas of texture modeling outside X-ray mammography as well, provide ideal visual verification using synthesis of the corresponding measured data spaces, contrary to stan- dard discriminative models. All achieved results in the thesis are extensively benchmarked. The thesis presents two methods for breast density classification in X-ray mammography. The methods were tested on the widely known MIAS database and the state-of-the art INbreast database, with competitive results. Several methods for completely automatic mammogram texture enhance- ment are presented. These methods are based on the descriptive textural mod- els developed in the thesis which automatically adapt to the analyzed X-ray texture, thus being universal for any type of input without the need of further manual tuning of specific parameters. The methods' outputs highlight regions of interest, detected as textural abnormalities. The methods provide the pos- sibility of enhancement tuned to specific types of mammogram tissue. Hence, the enhanced mammograms can help radiologists to decrease their false negative...
|
|
|
Query by Pictorial Example
Vácha, Pavel ; Haindl, Michal (advisor)
Ongoing expansion of digital images requires new methods for sorting, browsing, and sear- ching through huge image databases. This is a domain of Content-Based Image Retrieval (CBIR) systems, which are database search engines for images. A user typically submit a query image or series of images and the CBIR system tries to find and to retrieve the most similar images from the database. Optimally, the retrieved images should not be sensitive to circumstances during their acquisition. Unfortunately, the appearance of natural objects and materials is highly illumination and viewpoint dependent. This work focuses on representation and retrieval of homogeneous images, called textu- res, under the circumstances with variable illumination and texture rotation. We propose a novel illumination invariant textural features based on Markovian modelling of spatial tex- ture relations. The texture is modelled by Causal Autoregressive Random field (CAR) or Gaussian Markov Random Field (GMRF) models, which allow a very efficient estimation of its parameters, without the demanding Monte Carlo minimisation. Subsequently, the estimated model parameters are transformed into the new illumination invariants, which represent the texture. We derived that our textural representation is invariant to changes of illumination intensity and...
|
|
|
Query by Pictorial Example
Vácha, Pavel ; Haindl, Michal (advisor) ; Drbohlav, Ondřej (referee) ; Geusebroek, Jan-Mark (referee)
Ongoing expansion of digital images requires new methods for sorting, browsing, and sear- ching through huge image databases. This is a domain of Content-Based Image Retrieval (CBIR) systems, which are database search engines for images. A user typically submit a query image or series of images and the CBIR system tries to find and to retrieve the most similar images from the database. Optimally, the retrieved images should not be sensitive to circumstances during their acquisition. Unfortunately, the appearance of natural objects and materials is highly illumination and viewpoint dependent. This work focuses on representation and retrieval of homogeneous images, called textu- res, under the circumstances with variable illumination and texture rotation. We propose a novel illumination invariant textural features based on Markovian modelling of spatial tex- ture relations. The texture is modelled by Causal Autoregressive Random field (CAR) or Gaussian Markov Random Field (GMRF) models, which allow a very efficient estimation of its parameters, without the demanding Monte Carlo minimisation. Subsequently, the estimated model parameters are transformed into the new illumination invariants, which represent the texture. We derived that our textural representation is invariant to changes of illumination intensity and...
|
|
|
Image Segmentation
Mikeš, Stanislav ; Haindl, Michal (advisor) ; Scarpa, Giuseppe (referee) ; Jan, Jiří (referee)
Image segmentation is a fundamental part in low level computer vision processing. It has an essential influence on the subsequent higher level visual scene interpretation for a wide range of applications. Unsupervised image segmentation is an ill-defined problem and thus cannot be optimally solved in general. Several novel unsupervised multispectral image segmentation methods based on the underlaying random field texture models (GMRF, 2D/3D CAR) were developed. These segmenters use efficient data representations that allow an analytical solutions and thus the segmentation algorithm is much faster in comparison to methods based on MCMC. All segmenters were extensively compared with the alternative state- of-the-art segmenters with very good results. The MW3AR segmenter scored as one of the best available. The cluster validation problem was solved by a modified EM algorithm. Two multiple resolution segmenters were designed as a combination of a set of single segmenters. To tackle a realistic variable lighting in images, the illumination invariant features were derived and the illumination invariant segmenter was developed. For the proper evaluation of segmentation results and ranking of algorithms, a unique web-based texture segmentation benchmark was proposed and implemented. It was used for comprehensive...
|
|
|
Český filmový trh a digitální filmové pirátství: Ekonomická analýza a navržené politiky
Janák, Pavel ; Voráček, Jan (advisor) ; Haindl, Michal (referee) ; Jánský, Jaroslav (referee) ; Choochote, Kitimaporn (referee)
The Czech theatrical market faces a major digital piracy problem. The availability of illegitimate digital distribution channels represents a challenge for managers, especially when original movies are uploaded to the Internet before or during their theatrical release. The crucial managerial task is to solve the problems of losses caused by piracy, and to find a balance between antipiracy investment and the maximal benefits it brings. Therefore, to maximize stakeholder utility, management decision-making needs to be complemented and supported. Firstly, the research investigates the basis of digital piracy, discusses the effects it causes, focuses on relevant stakeholders. The thesis deals with knowledge management and system dynamics using its principles and approaches and proposes a model supporting strategic management decision-making. The created knowledge-based computational model simulates the market's development of the Czech theatrical industry related to digital film piracy in the following scenarios: current market settings, industry-based administration, government-based administration and a mixture of the last two; the results of the different scenarios are discussed, evaluated and compared. The findings indicate that with the current settings in the Czech theatrical industry, the losses caused by digital piracy keep increasing linearly. Industry-based administration implies that the losses level off with a negligible yearly increase and the government-based solution reduces the losses more than the industrial administration. Nevertheless, the results show minor differences in total box office revenues, while differences in antipiracy costs are vast. Therefore, the predictive experiment based on the current market environment represents the most efficient version of the experiments. Even though losses are, the highest, real box office revenues are only a little different. Simply put, a moderate increase in box office revenues paid for by massive investments into antipiracy seems inefficient.
|
guest ::
