Faculty of Electrical and Electronic Engineering and AutomationBook 3.2 · Communication and Computer Technologies

FRI-2G.302-1-CCT2-01

IMAGE RECOGNITION BY DISCRETE WAVELET TRANSFORM AND DECISION TREE METHOD

Assoc. Prof. Ivelina Balabanova, PhD

Department of Communications Equipment and Technologies Technical University of Gabrovo, Bulgaria

E-mail: ivstoeva@abv.bg

Chief Assist. Prof. Georgi Georgiev, PhD

Department of Communications Equipment and Technologies Technical University of Gabrovo, Bulgaria

E-mail: givanow@abv.bg

Eng. Kristina Sidorova, PhD Student

Department of Communications Equipment and Technologies Technical University of Gabrovo, Bulgaria

E-mail: kris_1981@mail.bg

Abstract

The paper summarizes the performance analysis results of a developed approach for processing and recognizing biometric facial images using Discrete Wavelet Transform (DWT) and Decision Tree (DT) techniques. Haar, Coiflets, Biorthogonal and Symplets wavelet feature extraction procedures were performed using a modeled GUI modeled in LabVIEW platform. Based on the DWTs applications, a complex of Approximation and Detail coefficients were obtained. Immediately before the training phase of DT-based classification models, an importance assessment and ranking of the included wavelet coefficients in the input dataset was conducted. Cross-validation and Resubstitution analytical techniques were applied in the performance evaluation phase of the generated categories Decision trees in reduction pruning levels in STATITICA software. An optimal architecture for multivariate decision making “Tree 4” with the most suitable parameterization for recognition of the test facial images with an accuracy of 91.325 % has been established.

Keywords

Face Recognition, Wavelet Transform, CART Training, Decision Tree, Cross-Validation, Resubstitution

REFERENCES

Ersoy, E., Albey, E., & Kayis, E. (2020). A CART-based Genetic Algorithm for Constructing Higher Accuracy Decision Trees. XI<sup>th</sup> International Conference on Data Science, Technology and Applications, 7th-9th July 2020, Online Streaming.
Gupta, B., Rawat, A., Jain, A., Arora, A., & Dhami, N. (2017). Analysis of Various Decision Tree Algorithms for Classification and Data Mining. International Journal of Computer Applications, 163(8), pp. 15-19.
Huynh-Cam, T., Chen, L., & Le, H. (2021). Using Decision Trees and Random Forest Algorithms to Predict and Determinate Factors Contributing to First-year University Students’ Learning Performance. MDPI Algorithms, 14(318), pp. 1-17.
Mienye, I., & Jere, N. (2016). A Survey of Decision Trees: Concepts, Algorithms and Applications. IEEE Access, vol. 4, pp. 1-15.
Mittal, K., Khanduja, D., & Tewari, P. (2017). An Insight into Decision Tree Analysis. World Wide Journal of Multidisciplinary Research and Development, 3(12), pp. 111-115.
Shah, T., Khan, M., Ali, M., Khan, B., & Idress, N. (2018). CART, J-48graft, J48, Decision Stump and Random Forest: a Comparative Study. University of Swabi Journal, 2(1), pp. 1-6.
Sharma, H., & Kumar, S. (2013). A Survey on Decision Tree Algorithms of Classification in Data Mining. International Journal of Science and Research, 5(4), pp. 2094-2097.
Talekar, B., & Agrawal, S. (2020). A Detailed Review on Decision Tree and Random Forest. Bioscience Biotechnology, Research Communications, 13(14), pp. 245-248.
Zhang, Z. (2021). Applications of the Decision Tree in Business Field. Advances in Economics, Business and Management Research, 203, pp. 926-929.