Information-extremе machine learning for object identification on the terrain
The article deals with the usage of a SURF local descriptor of key fragments to create a global descriptor BoF for objects of interest on terrain within task of recognition of armored technique in the controlled territory using images of air reconnaissance. The method of optimization of the input ma...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Sumy State University
2016-06-01
|
| Series: | Журнал інженерних наук |
| Subjects: | |
| Online Access: | http://jes.sumdu.edu.ua/wp-content/uploads/2016/08/JES_2016_01_H_1_V3.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850178853533646848 |
|---|---|
| author | V. V. Moskalenko A. H. Korobov |
| author_facet | V. V. Moskalenko A. H. Korobov |
| author_sort | V. V. Moskalenko |
| collection | DOAJ |
| description | The article deals with the usage of a SURF local descriptor of key fragments to create a global descriptor BoF for objects of interest on terrain within task of recognition of armored technique in the controlled territory using images of air reconnaissance. The method of optimization of the input mathematical description of the information-extreme classifier trained on dataset, which consist of global descriptors BoF, is proposed. The optimal in information understanding dimensionality of a global descriptor is determined by iterative procedure, which includes k-means clustering of key fragment's SURF-vectors, training dataset creation and information-extreme machine learning of the classifier. The offered algorithm of information-extreme machine learning implements the adaptive coding of values of primary features using multilevel system of control permits, and creation of hyperspherical containers of classes in binary space of secondary features with sequential optimization procedures. It was suggested to use the rated modification of S. Kulbak's information measure, which is a function of false omission rate and positive predictive value of decision-making and it also allows machine learning on imbalanced dataset. |
| format | Article |
| id | doaj-art-c5e7bd218cb04c9c99a42daa27790da8 |
| institution | OA Journals |
| issn | 2312-2498 2414-9381 |
| language | English |
| publishDate | 2016-06-01 |
| publisher | Sumy State University |
| record_format | Article |
| series | Журнал інженерних наук |
| spelling | doaj-art-c5e7bd218cb04c9c99a42daa27790da82025-08-20T02:18:38ZengSumy State UniversityЖурнал інженерних наук2312-24982414-93812016-06-0131H1H7Information-extremе machine learning for object identification on the terrainV. V. Moskalenko0 A. H. Korobov1Sumy State UniversitySumy State UniversityThe article deals with the usage of a SURF local descriptor of key fragments to create a global descriptor BoF for objects of interest on terrain within task of recognition of armored technique in the controlled territory using images of air reconnaissance. The method of optimization of the input mathematical description of the information-extreme classifier trained on dataset, which consist of global descriptors BoF, is proposed. The optimal in information understanding dimensionality of a global descriptor is determined by iterative procedure, which includes k-means clustering of key fragment's SURF-vectors, training dataset creation and information-extreme machine learning of the classifier. The offered algorithm of information-extreme machine learning implements the adaptive coding of values of primary features using multilevel system of control permits, and creation of hyperspherical containers of classes in binary space of secondary features with sequential optimization procedures. It was suggested to use the rated modification of S. Kulbak's information measure, which is a function of false omission rate and positive predictive value of decision-making and it also allows machine learning on imbalanced dataset.http://jes.sumdu.edu.ua/wp-content/uploads/2016/08/JES_2016_01_H_1_V3.pdfmachine learningdescriptorunmanned aerial vehicleclassfeature setinformation criterionoptimizations |
| spellingShingle | V. V. Moskalenko A. H. Korobov Information-extremе machine learning for object identification on the terrain Журнал інженерних наук machine learning descriptor unmanned aerial vehicle class feature set information criterion optimization s |
| title | Information-extremе machine learning for object identification on the terrain |
| title_full | Information-extremе machine learning for object identification on the terrain |
| title_fullStr | Information-extremе machine learning for object identification on the terrain |
| title_full_unstemmed | Information-extremе machine learning for object identification on the terrain |
| title_short | Information-extremе machine learning for object identification on the terrain |
| title_sort | information extremе machine learning for object identification on the terrain |
| topic | machine learning descriptor unmanned aerial vehicle class feature set information criterion optimization s |
| url | http://jes.sumdu.edu.ua/wp-content/uploads/2016/08/JES_2016_01_H_1_V3.pdf |
| work_keys_str_mv | AT vvmoskalenko informationextrememachinelearningforobjectidentificationontheterrain AT ahkorobov informationextrememachinelearningforobjectidentificationontheterrain |