ON PERFORMANCE INDICATORS OF RECOGNITION METHODS, PROVIDED THAT THE TEST DATA IS APPROXIMATELY MARKED UP
Various performance indicators are used to evaluate and compare the effectiveness of object recognition methods in solving a specific task. When evaluating the performance of these methods based on the analysis of real test data, the operator's marking of pixels belonging to an object can in many cases be performed fairly approximately. The paper proposes estimates of Accuracy, Precision, Recall, and F1-score indicators of the efficiency of recognition (classification) methods under the condition of approximate markup of test data (images). The paper also suggests the so-called normalized mean-square distance between a set of False Positive pixels and a set of pixels of an object as an indicator of the performance of recognition methods, which, unlike other indicators, allows us to estimate the distribution of False Positive pixels in an image relative to pixels of objects, which is important when evaluating and comparing the effectiveness of various recognition methods. The paper provides examples of calculating the values of the proposed indicators.
Bolgova E.V., Chernomorets A.A., Bukhantsov A.D., Funikov A.D. On Performance Indicators of Recognition Methods, Provided that the Test Data is Approximately Marked Up // Research result. Information technologies. – T.10, №4, 2025. – P. 3-13. DOI: 10.18413/2518-1092-2025-10-4-0-1
While nobody left any comments to this publication.
You can be first.
1. Mikhailichenko A.A. An Analytical Review of Methods for Assessing the Quality of Classification Algorithms in Machine Learning Problems. Bulletin of Adyghe State University. Series 4: Natural, Mathematical, and Technical Sciences. 2022. No. 4 (311). pp. 52–59.
2. Limanovskaya O.V., Alferyeva T.I. Fundamentals of Machine Learning: A Textbook. Yekaterinburg: Ural University Press, 2020. 88 p.
3. Levchuk S.A., Yakimenko A.A. A Study of the Characteristics of Face Recognition Algorithms. Collection of Scientific Papers of NSTU. 2018. No. 3–4 (93). pp. 40–58.
4. Lazarev D.A., Funikov A.D., Bolgova E.V., Chernomorets A.A., Fefelov O.S. 2025 On Algorithms for Segmentation of Digital Images of Motor Roads. Economics. Information technologies, 52(1): 215-226 (in Russian). DOI 10.52575/2687-0932-2025-52-1-215-226.
5. Applied Mathematical Statistics: A Textbook. Author: A.A. Mitsel. – Tomsk: Tomsk State University of Control Systems and Radioelectronics, 2019. – 113 p.
6. Serra J. Image Analysis and Mathematical Morphology. – 1982. – 610 p.
7. Serra J. Image Analysis and Mathematical Morphology. Vol. 2: Theoretical Advances. – 1988. – 411 p.
8. Gonzalez R. Digital Image Processing. 3rd edition, revised and supplemented / R. Gonzalez, R. Woods. – Moscow: Tekhnosfera, 2012. – 1104 p.
9. Lebedev L.I., Vasin Yu.G. A two-criteria algorithm for recognizing objects in graphic images based on the KECM // 25th Anniversary International Conference (GraphiCon2015), Russia, Protvino (Park Drakino), September 22–25, 2015. – P. 112–114.
10. Khmelev R.V. Joint use of structural analysis and the Hausdorff metric in comparing an object and a standard // Computer Optics, 2005, No. 27 – P. 174–176.
11. Hausdorff F. Set Theory. – M.-L.: United Scientific and Technical Publishing House of the People's Commissariat of Industrial Trade of the USSR, 1937. – 305 p.
12. Sukharev A.G., Timokhov A.V., Fedorov V.V. Course of optimization methods: Textbook. – 2nd ed. – Moscow: Fizmatlit, 2011. – 384 p.