2016 |
Wilkowski, Artur; Luckner, Marcin Low-cost canoe counting system for application in a natural environment Inproceedings Advances in Intelligent Systems and Computing, pp. 705–715, 2016, ISSN: 21945357. Abstract | Links | BibTeX | Tags: Classification with rejection, Computer vision, Pattern recognition @inproceedings{Wilkowski2016, title = {Low-cost canoe counting system for application in a natural environment}, author = {Artur Wilkowski and Marcin Luckner}, doi = {10.1007/978-3-319-29357-8_61}, issn = {21945357}, year = {2016}, date = {2016-01-01}, booktitle = {Advances in Intelligent Systems and Computing}, volume = {440}, pages = {705--715}, abstract = {? Springer International Publishing Switzerland 2016.This paper presents low-cost system for counting canoes and canoeists to control cannoning tourist routes. The created system was implemented on Raspberry Pi 2 and the total cost of the tracking device is less than 200$. The proposed algorithmuses background subtraction and Support Vector Machines to track vessels and recognize canoes among them. The obtained results are rewarding as for low-cost solution. Depending on considered group of objects the accuracy of the algorithm reaches 84, 89.5, and 96% for canoes, vessels, and all objects respectively.}, keywords = {Classification with rejection, Computer vision, Pattern recognition}, pubstate = {published}, tppubtype = {inproceedings} } ? Springer International Publishing Switzerland 2016.This paper presents low-cost system for counting canoes and canoeists to control cannoning tourist routes. The created system was implemented on Raspberry Pi 2 and the total cost of the tracking device is less than 200$. The proposed algorithmuses background subtraction and Support Vector Machines to track vessels and recognize canoes among them. The obtained results are rewarding as for low-cost solution. Depending on considered group of objects the accuracy of the algorithm reaches 84, 89.5, and 96% for canoes, vessels, and all objects respectively. |
Wilkowski, Artur; Luckner, Marcin Low-cost canoe counting system for application in a natural environment Inproceedings Advances in Intelligent Systems and Computing, pp. 705–715, 2016, ISSN: 21945357. Abstract | Links | BibTeX | Tags: Classification with rejection, Computer vision, Pattern recognition @inproceedings{Wilkowski2016b, title = {Low-cost canoe counting system for application in a natural environment}, author = {Artur Wilkowski and Marcin Luckner}, doi = {10.1007/978-3-319-29357-8_61}, issn = {21945357}, year = {2016}, date = {2016-01-01}, booktitle = {Advances in Intelligent Systems and Computing}, volume = {440}, pages = {705--715}, abstract = {? Springer International Publishing Switzerland 2016.This paper presents low-cost system for counting canoes and canoeists to control cannoning tourist routes. The created system was implemented on Raspberry Pi 2 and the total cost of the tracking device is less than 200$. The proposed algorithmuses background subtraction and Support Vector Machines to track vessels and recognize canoes among them. The obtained results are rewarding as for low-cost solution. Depending on considered group of objects the accuracy of the algorithm reaches 84, 89.5, and 96% for canoes, vessels, and all objects respectively.}, keywords = {Classification with rejection, Computer vision, Pattern recognition}, pubstate = {published}, tppubtype = {inproceedings} } ? Springer International Publishing Switzerland 2016.This paper presents low-cost system for counting canoes and canoeists to control cannoning tourist routes. The created system was implemented on Raspberry Pi 2 and the total cost of the tracking device is less than 200$. The proposed algorithmuses background subtraction and Support Vector Machines to track vessels and recognize canoes among them. The obtained results are rewarding as for low-cost solution. Depending on considered group of objects the accuracy of the algorithm reaches 84, 89.5, and 96% for canoes, vessels, and all objects respectively. |
2014 |
Homenda, Wladyslaw; Luckner, Marcin Pattern recognition with rejection: Application to handwritten digits Inproceedings 2014 4th World Congress on Information and Communication Technologies (WICT 2014), pp. 326–331, IEEE, 2014, ISBN: 978-1-4799-8115-1. Abstract | Links | BibTeX | Tags: Accuracy, Classification with rejection, Handwriting recognition, native and foreign elements, pattern recognition with rejection, Standards, Support Vector Machines, Testing, Text recognition @inproceedings{Homenda2014a, title = {Pattern recognition with rejection: Application to handwritten digits}, author = {Wladyslaw Homenda and Marcin Luckner}, url = {http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7077288}, doi = {10.1109/WICT.2014.7077288}, isbn = {978-1-4799-8115-1}, year = {2014}, date = {2014-12-01}, booktitle = {2014 4th World Congress on Information and Communication Technologies (WICT 2014)}, pages = {326--331}, publisher = {IEEE}, abstract = {The paper considers rejecting option in pattern recognition problem. Studied are native and foreign elements in a multi-class pattern recognition. Native elements are those included in recognized classes, they are known at the stage of classifier design. Foreign elements do not belong to recognized classes. Usually foreign elements are not known when classifier is designed. If foreign elements are classified to recognized classes, recognition quality is deteriorated. So then, they are classified to native classes, if they are not rejected. In such the case, recognition quality is deteriorated. Therefore, they should be rejected by a classifier, i.e. not classified to any class. Several attempts to rejection of foreign elements are investigated in this study.}, keywords = {Accuracy, Classification with rejection, Handwriting recognition, native and foreign elements, pattern recognition with rejection, Standards, Support Vector Machines, Testing, Text recognition}, pubstate = {published}, tppubtype = {inproceedings} } The paper considers rejecting option in pattern recognition problem. Studied are native and foreign elements in a multi-class pattern recognition. Native elements are those included in recognized classes, they are known at the stage of classifier design. Foreign elements do not belong to recognized classes. Usually foreign elements are not known when classifier is designed. If foreign elements are classified to recognized classes, recognition quality is deteriorated. So then, they are classified to native classes, if they are not rejected. In such the case, recognition quality is deteriorated. Therefore, they should be rejected by a classifier, i.e. not classified to any class. Several attempts to rejection of foreign elements are investigated in this study. |
Homenda, Wladyslaw; Luckner, Marcin Pattern recognition with rejection: Application to handwritten digits Inproceedings 2014 4th World Congress on Information and Communication Technologies (WICT 2014), pp. 326–331, IEEE, 2014, ISBN: 978-1-4799-8115-1. Abstract | Links | BibTeX | Tags: Accuracy, Classification with rejection, Handwriting recognition, native and foreign elements, pattern recognition with rejection, Standards, Support Vector Machines, Testing, Text recognition @inproceedings{Homenda2014ab, title = {Pattern recognition with rejection: Application to handwritten digits}, author = {Wladyslaw Homenda and Marcin Luckner}, url = {http://ieeexplore.ieee.org/lpdocs/epic03/wrapper.htm?arnumber=7077288}, doi = {10.1109/WICT.2014.7077288}, isbn = {978-1-4799-8115-1}, year = {2014}, date = {2014-12-01}, booktitle = {2014 4th World Congress on Information and Communication Technologies (WICT 2014)}, pages = {326--331}, publisher = {IEEE}, abstract = {The paper considers rejecting option in pattern recognition problem. Studied are native and foreign elements in a multi-class pattern recognition. Native elements are those included in recognized classes, they are known at the stage of classifier design. Foreign elements do not belong to recognized classes. Usually foreign elements are not known when classifier is designed. If foreign elements are classified to recognized classes, recognition quality is deteriorated. So then, they are classified to native classes, if they are not rejected. In such the case, recognition quality is deteriorated. Therefore, they should be rejected by a classifier, i.e. not classified to any class. Several attempts to rejection of foreign elements are investigated in this study.}, keywords = {Accuracy, Classification with rejection, Handwriting recognition, native and foreign elements, pattern recognition with rejection, Standards, Support Vector Machines, Testing, Text recognition}, pubstate = {published}, tppubtype = {inproceedings} } The paper considers rejecting option in pattern recognition problem. Studied are native and foreign elements in a multi-class pattern recognition. Native elements are those included in recognized classes, they are known at the stage of classifier design. Foreign elements do not belong to recognized classes. Usually foreign elements are not known when classifier is designed. If foreign elements are classified to recognized classes, recognition quality is deteriorated. So then, they are classified to native classes, if they are not rejected. In such the case, recognition quality is deteriorated. Therefore, they should be rejected by a classifier, i.e. not classified to any class. Several attempts to rejection of foreign elements are investigated in this study. |
Publications
2016 |
Low-cost canoe counting system for application in a natural environment Inproceedings Advances in Intelligent Systems and Computing, pp. 705–715, 2016, ISSN: 21945357. |
Low-cost canoe counting system for application in a natural environment Inproceedings Advances in Intelligent Systems and Computing, pp. 705–715, 2016, ISSN: 21945357. |
2014 |
Pattern recognition with rejection: Application to handwritten digits Inproceedings 2014 4th World Congress on Information and Communication Technologies (WICT 2014), pp. 326–331, IEEE, 2014, ISBN: 978-1-4799-8115-1. |
Pattern recognition with rejection: Application to handwritten digits Inproceedings 2014 4th World Congress on Information and Communication Technologies (WICT 2014), pp. 326–331, IEEE, 2014, ISBN: 978-1-4799-8115-1. |