Marcin Iwanowski - Academia.edu (original) (raw)
Uploads
Papers by Marcin Iwanowski
Future Generation Computer Systems
Annals of Computer Science and Information Systems
Computer Science Research Notes, 2021
Geoscience data journal, Jun 1, 2017
Lecture notes in networks and systems, 2023
Advances in intelligent systems and computing, Oct 15, 2015
Multiscale feature point descriptors are used to describe the surrounding of image feature points... more Multiscale feature point descriptors are used to describe the surrounding of image feature points, taking into account neighboring image details at various levels. The feature vectors are divided into several parts, each of which describes surrounding at increasing distance from the feature point. In this paper a method for matching such descriptors is proposed. It is based on multiple nearest-neighbors searches that are applied to match parts of the descriptor, followed by combining the partial matching results into final indication of the closest descriptor.
Springer eBooks, Sep 28, 2017
In the paper, the concept of a system is presented, that makes use of the reasoning to perform di... more In the paper, the concept of a system is presented, that makes use of the reasoning to perform digital image search using the information present in an annotated images combined with external text data sources in order to answer complex queries. The workflow in based the Prolog reasoning that allows for both modeling ontologies and combining the knowledge from both data sources. Prior to reasoning, both data sources are transformed into the form of Prolog predicates. An example application of the proposed concept is also presented in the paper. It deals with photographs database with annotated faces of politicians combined with the external knowledge stored in the Parliament-members database. It provides with a mechanism that supports formulating complex questions like: “Find a photograph where a politician belonging to party A stands between a woman and another political that was a member of Parliament of B-th term of office”.
Lecture Notes in Computer Science, 2018
The paper presents simple graph features based on a well-known image keypoints. We discuss the ex... more The paper presents simple graph features based on a well-known image keypoints. We discuss the extraction method and geometrical properties that can be used. Chosen methods are tested in KNN tasks for almost 1000 object classes. The approach addresses problems in applications that cannot use learning methods explicitly, as real-time tracking, chosen object detection scenarios and structure from motion. Results imply that the idea is worth further research for chosen systems.
Advances in intelligent systems and computing, May 7, 2017
Search engines have become one of indispensable elements of modern information processing systems... more Search engines have become one of indispensable elements of modern information processing systems. One of challenges to face in this area is to build an interface between pictorial and text data, that would allow for image search using text queries. In the paper, a concept is proposed, that combines inference, ontology and natural language processing to formulate queries to the pictorial data of raster map images. The proposed system analyses a user’s query formulated in the natural language and finds place on map which meet this criteria. The proposed approach improves remarkably the easy of access to the raster map content.
Advances in intelligent systems and computing, May 7, 2017
Machine graphics & vision, 2001
Machine graphics & vision, 2001
Advances in Intelligent Systems and Computing, 2019
In the paper, a method of contrast enhancement is presented. It combines linear and non-linear te... more In the paper, a method of contrast enhancement is presented. It combines linear and non-linear techniques. A non-linear approach based on morphological reconstruction is applied to select essential regions on the of Laplacian-of-Gaussian (LoG) of the input. The morphologically modified result of LoG is next added to the initial image. Thanks to the morphological processing, only the meaningful region boundaries are used to enhance the image contrast. The usage of morphological processing allows for selecting more precisely the image regions that are subject to contrast enhancement. The method performs well on textured images, allowing for adjusting the level of visible details in the output image while increasing the sharpness of meaningful image regions. A couple of examples illustrates the performance of the proposed method.
Future Generation Computer Systems
Annals of Computer Science and Information Systems
Computer Science Research Notes, 2021
Geoscience data journal, Jun 1, 2017
Lecture notes in networks and systems, 2023
Advances in intelligent systems and computing, Oct 15, 2015
Multiscale feature point descriptors are used to describe the surrounding of image feature points... more Multiscale feature point descriptors are used to describe the surrounding of image feature points, taking into account neighboring image details at various levels. The feature vectors are divided into several parts, each of which describes surrounding at increasing distance from the feature point. In this paper a method for matching such descriptors is proposed. It is based on multiple nearest-neighbors searches that are applied to match parts of the descriptor, followed by combining the partial matching results into final indication of the closest descriptor.
Springer eBooks, Sep 28, 2017
In the paper, the concept of a system is presented, that makes use of the reasoning to perform di... more In the paper, the concept of a system is presented, that makes use of the reasoning to perform digital image search using the information present in an annotated images combined with external text data sources in order to answer complex queries. The workflow in based the Prolog reasoning that allows for both modeling ontologies and combining the knowledge from both data sources. Prior to reasoning, both data sources are transformed into the form of Prolog predicates. An example application of the proposed concept is also presented in the paper. It deals with photographs database with annotated faces of politicians combined with the external knowledge stored in the Parliament-members database. It provides with a mechanism that supports formulating complex questions like: “Find a photograph where a politician belonging to party A stands between a woman and another political that was a member of Parliament of B-th term of office”.
Lecture Notes in Computer Science, 2018
The paper presents simple graph features based on a well-known image keypoints. We discuss the ex... more The paper presents simple graph features based on a well-known image keypoints. We discuss the extraction method and geometrical properties that can be used. Chosen methods are tested in KNN tasks for almost 1000 object classes. The approach addresses problems in applications that cannot use learning methods explicitly, as real-time tracking, chosen object detection scenarios and structure from motion. Results imply that the idea is worth further research for chosen systems.
Advances in intelligent systems and computing, May 7, 2017
Search engines have become one of indispensable elements of modern information processing systems... more Search engines have become one of indispensable elements of modern information processing systems. One of challenges to face in this area is to build an interface between pictorial and text data, that would allow for image search using text queries. In the paper, a concept is proposed, that combines inference, ontology and natural language processing to formulate queries to the pictorial data of raster map images. The proposed system analyses a user’s query formulated in the natural language and finds place on map which meet this criteria. The proposed approach improves remarkably the easy of access to the raster map content.
Advances in intelligent systems and computing, May 7, 2017
Machine graphics & vision, 2001
Machine graphics & vision, 2001
Advances in Intelligent Systems and Computing, 2019
In the paper, a method of contrast enhancement is presented. It combines linear and non-linear te... more In the paper, a method of contrast enhancement is presented. It combines linear and non-linear techniques. A non-linear approach based on morphological reconstruction is applied to select essential regions on the of Laplacian-of-Gaussian (LoG) of the input. The morphologically modified result of LoG is next added to the initial image. Thanks to the morphological processing, only the meaningful region boundaries are used to enhance the image contrast. The usage of morphological processing allows for selecting more precisely the image regions that are subject to contrast enhancement. The method performs well on textured images, allowing for adjusting the level of visible details in the output image while increasing the sharpness of meaningful image regions. A couple of examples illustrates the performance of the proposed method.