“…2) By assigning individually each of the object descriptors to one of the etalons with the determination of the etalon that received the largest number of class votes [7], [11], [37]. 3) Using the apparatus of statistical data analysis to make decision H about a significant difference between the distributions of data for describing the object and etalons with the definition of the etalon, which has a non-significant difference [8], [33], [35]. With this approach, according to the apparatus of statistical testing of hypothesis )] , ( ), , ( […”
Section: Methods Of Constructing Classifiers According To Transformed...mentioning
confidence: 99%
“…The variants of implementing the third method are implemented, for example, by comparing the distributions of object data and etalons through the application of a twosample t -test for meansboth for bit-by-bit comparison of the vectors, which are integrated representations of the objects and for comparing such vectors as a whole [33], [35].…”
Section: Methods Of Constructing Classifiers According To Transformed...mentioning
confidence: 99%
“…1) The construction and use of data distributions instead of data values significantly reduce the number of calculations [2], [33].…”
Section: Review Of the Literaturementioning
confidence: 99%
“…Such approaches by generalizing data transformation significantly simplify and speed up classification. At the same time, it is effective to combine methods and implement a complex of approaches to improve performance, namely: reducing the number of descriptors in the description [37], statistical processing with the identification of the most informative components [16], [29], [30], [33], [35], [37]. The possibility of implementing training procedures in such classifiers, where etalon descriptions of classes are considered to be given, also contributes to further improvement of their performance.…”
The aim of the research is to improve the effectiveness of image recognition methods according to the description in the form of a set of keypoint descriptors. The focus is on increasing the speed of analysis and processing of description data while maintaining the required level of classification efficiency. The class of the image is provided as a description of the etalon. It is proposed to transform the description by implementing a statistical system of features for non-intersecting data fragments. The developed method is based on the aggregation of data distribution values within the description, the basis of which is the bit representation of the descriptors. Statistical features are calculated as the frequency of occurrence of the fixed value of a fragment on a set of description data and thus reflect the individual properties of images. Three main classifier models are analyzed: calculating the measure of data relevance in the form of distributions; assigning each of the descriptors to defined classes (voting); using the apparatus of statistical data analysis to decide on the significance of the difference between the distributions of the object and etalons. The results of software modeling of methods and calculations of statistical significance of differences based on distributions for training sets of images are represented. Using distributions instead of a set of descriptors increases the processing speed by hundreds of times, while the classification accuracy is maintained at a sufficient level and does not deteriorate compared to traditional voting.INDEX TERMS Computer vision, data fragment, descriptor, image classification, keypoint.
“…2) By assigning individually each of the object descriptors to one of the etalons with the determination of the etalon that received the largest number of class votes [7], [11], [37]. 3) Using the apparatus of statistical data analysis to make decision H about a significant difference between the distributions of data for describing the object and etalons with the definition of the etalon, which has a non-significant difference [8], [33], [35]. With this approach, according to the apparatus of statistical testing of hypothesis )] , ( ), , ( […”
Section: Methods Of Constructing Classifiers According To Transformed...mentioning
confidence: 99%
“…The variants of implementing the third method are implemented, for example, by comparing the distributions of object data and etalons through the application of a twosample t -test for meansboth for bit-by-bit comparison of the vectors, which are integrated representations of the objects and for comparing such vectors as a whole [33], [35].…”
Section: Methods Of Constructing Classifiers According To Transformed...mentioning
confidence: 99%
“…1) The construction and use of data distributions instead of data values significantly reduce the number of calculations [2], [33].…”
Section: Review Of the Literaturementioning
confidence: 99%
“…Such approaches by generalizing data transformation significantly simplify and speed up classification. At the same time, it is effective to combine methods and implement a complex of approaches to improve performance, namely: reducing the number of descriptors in the description [37], statistical processing with the identification of the most informative components [16], [29], [30], [33], [35], [37]. The possibility of implementing training procedures in such classifiers, where etalon descriptions of classes are considered to be given, also contributes to further improvement of their performance.…”
The aim of the research is to improve the effectiveness of image recognition methods according to the description in the form of a set of keypoint descriptors. The focus is on increasing the speed of analysis and processing of description data while maintaining the required level of classification efficiency. The class of the image is provided as a description of the etalon. It is proposed to transform the description by implementing a statistical system of features for non-intersecting data fragments. The developed method is based on the aggregation of data distribution values within the description, the basis of which is the bit representation of the descriptors. Statistical features are calculated as the frequency of occurrence of the fixed value of a fragment on a set of description data and thus reflect the individual properties of images. Three main classifier models are analyzed: calculating the measure of data relevance in the form of distributions; assigning each of the descriptors to defined classes (voting); using the apparatus of statistical data analysis to decide on the significance of the difference between the distributions of the object and etalons. The results of software modeling of methods and calculations of statistical significance of differences based on distributions for training sets of images are represented. Using distributions instead of a set of descriptors increases the processing speed by hundreds of times, while the classification accuracy is maintained at a sufficient level and does not deteriorate compared to traditional voting.INDEX TERMS Computer vision, data fragment, descriptor, image classification, keypoint.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.