LOGICAL ANALYSIS AND PROCESSING OF DATA FOR THE CLASSIFICATION OF IMAGES ON THE BASIS OF FORMATION OF STATISTICAL CENTER DESCRIPTION
DOI:
https://doi.org/10.26906/SUNZ.2019.4.043Keywords:
structural image recognition methods, key point, BRISK detector, descriptor, statistical center, concentrated description, logical analysis, relevance of descriptions, the effectiveness of classification, aggregated imageAbstract
The subjects of research are the models for classifying images in the description space as multiple key point descriptors when recognizing visual objects in computer vision systems. The goal is to develop a structural method of classification by introducing logical data processing using probability distribution in the form of statistical center. The tasks include the development of mathematical and software models to calculate the relevance of image descriptions using logical analysis, study of properties, variations of application, values of model parameters, evaluation of the results of processing the experimental image database. The methods are used: a BRISK detector for forming the key point descriptors, data mining, mathematical statistics, means of determining relevance for data sets, software modeling. The following results were obtained. The effectiveness of a method of classification based on logical analysis using statistical centers that depend on the distances between centers of etalons. Logical analysis simplifies processing and increases classification speed. The best results on the classification of individual descriptors were shown by the use of refined centers. Using a concentrated portion of the description data allows you to focus more closely on its differences with other descriptions. Conclusions. The contribution of the paper is the improvement of the image classification based on the implementation of logical analysis of the statistical center description, which allows modifying the composition of the description while maintaining the properties of objects in terms of effective classification. The practical significance of the paper is the achievement of the accepted relevance level according to a defined model of relevance, to validate the proposed modifications to the processing of sample images, to develop software models for implementing the described methods of classification in computer vision.Downloads
References
Xu Zhang, Felix X. Yu, Svebor Karaman, Shih-Fu Chang (2017). Learning Discriminative and Transformation Covariant Local Feature Detectors. − The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 6818-6826.
Masamichi KITAGAWA, Ikuko SHIMIZU (2019). Memory Saving Feature Descriptor Using Scale and Rotation Invariant Patches around the Feature Points. − IEICE Transactions on Information and Systems, Vol. E102.D, Issue 5, рр. 1106-1110.
Ravimal Bandara, Lochandaka Ranathunga, Nor Aniza Abdullah (2019). Nature Inspired Dimensional Reduction Technique for Fast and Invariant Visual Feature Extraction. − Int. Journal of Adv. Trends in Comp. Sc. Eng., Vol. 8, No.3, рр. 696-706.
Sivaram, M., Porkodi, V., Mohammed, A.S., Manikandan V. Detection of Accurate Facial Detection Using Hybrid Deep Convolutional Recurrent Neural Network. ICTACT Journal on Soft Computing. 2019. Vol. 09, Issue 02. pp. 1844-1850. DOI: 10.21917/ijsc.2019.0256
Amin Salih Mohammed, Saravana Balaji B., Hiwa Abdulkarim Mawlood. Conceptual analysis of Iris Recognition Systems. Advanced Information Systems. 2019. Vol. 3, No. 2. Р. 86-90. DOI : https://doi.org/10.20998/2522-9052.2019.2.15
Svyrydov, A., Kuchuk, H., Tsiapa, O. (2018), “Improving efficienty of image recognition process: Approach and case study”, Proceedings of 2018 IEEE 9th International Conference on Dependable Systems, Services and Technologies, DESSERT 2018, pp. 593-597, DOI: http://dx.doi.org/10.1109/DESSERT.2018.8409201
Худов В.Г. Аналіз відомих методів сегментування зображень, що отримані з бортових систем оптикоелектронного спостереження / В.Г. Худов, Г.А. Кучук, О.М. Маковейчук, А.В. Крижний // Системи обробки інформації, 2016. – Вип. 9 (146). – С. 77-80.
Yaloveha V., Hlavcheva D., Podorozhniak A. Usage of convolutional neural network for multispectral image processing applied to the problem of detecting fire hazardous forest areas. Сучасні інформаційні системи. 2019. Т. 3, № 1. С. 116– 120. DOI: https://doi.org/10.20998/2522-9052.2019.1.19.
Баклицкий, В.К. Методы фильтрации сигналов в корреляционно-экстремальных системах навигации / В.К. Баклицкий, А.М. Бочкарев, М.П. Мусьяков. – М.: Радио и связь, 1986. – 216 с.
Gorokhovatskyi, V., Gadetska, S., Ponomarenko, R. (2019) Recognition of Visual Objects Based on Statistical Distributions for Blocks of Structural Description of Image. Lecture Notes in Computational Intelligence and Decision Making. – Proc. of the XV Int. Scientific Conf. “Intellectual Systems of Decision Making and Problems of Computational Intelligence” (ISDMCI' 2019), Ukraine, May 21–25, 2019, pp. 501-512. – Available online: https://rd.springer.com/chapter/10.1007/978-3-030-26474-1_35
Gorokhovatskyi, O. Analysis of Application of Cluster Descriptions in Space of Characteristic Image Features / O. Gorok- hovatskyi, V. Gorokhovatskyi, O.Peredrii // Data. – 2018, 3(4), 52. – doi: 10.3390/data3040052. Available online: https://www.mdpi.com/2306-5729/3/4/52
Lyashenko V., Kobylin O., Baranchykov Y. Ideology of Image Processing in Infocommunication Systems // 2018 Int. Scientific-Practical Conference Problems of Infocommunications. Science and Technology (PIC S&T). – IEEE, 2018. – Р. 47-50.
Миркин, Б. Г. Методы кластер-анализа для поддержки принятия решений: обзор. Препринт WP7/2011/03 / Б. Г. Миркин. – М.: Изд. дом НИУ ВШЭ, 2011. – 88 с.
Гороховатський В.О. Статистичні розподіли та ланцюжкове подання даних при визначенні релевантності структурних описів візуальних об’єктів / В.О. Гороховатський, С.В. Гадецька, Р.П. Пономаренко // Системи управління, навігації та зв’язку. –2018. – №6 (52). – C. 87–92.
Гороховатський В.О. Вивчення статистичних властивостей моделі блочного подання для множини дескрипторів ключових точок зображень / В.О. Гороховатський , С.В. Гадецька, Н.І. Стяглик // Радіоелектроніка, інформатика, управління. –2019. – №2 . – C. 100–107.
Leutenegger, S., Chli, M., Siegwart, R. Y. (2011) BRISK: Binary Robust Invariant Scalable Keypoints. – Computer Vision (ICCV), pp. 2548 – 2555.
OpenCV Open Source Computer Vision. [Електронний ресурс]. – Режим доступу: https://docs.opencv.org/master/index.html
