A Deep Learning Approach for Classifying Generated Geometric Pattern Datasets
DOI:
https://doi.org/10.19139/soic-2310-5070-4116Keywords:
CNN Architecture, IGPs,, pre-trained models, ClassificationAbstract
Deep learning is now widely recognized as a powerful method for solving complex problems, such as imageclassification and object detection. This study focuses on applying deep learning to classify a custom-generated datasetof geometric patterns. We have implemented a binary classification system to distinguish between acceptable and non-acceptable geometric patterns based on criteria established by artisans working with wood. Due to the absence of a standarddataset, we created our own dataset, adhering to the construction rules for these patterns. In this study, two pre-trained CNN-driven models (Inception-V3 and ResNetV2) have been suggested in this paper for the classification. Numerous criteria,including precision, accuracy, recall, and F1 score, are used to illustrate this superiority. The experimental system focuses onevaluations based on loss, precision, and confusion matrices and uses Jupyter Notebook, Python, TensorFlow, and Keras. Thepre-trained Inception-V3 model achieves the best classification performance, with an accuracy of 97.80%, according to theexperimental results.Downloads
Published
2026-06-15
How to Cite
AIT LAHCEN, yassine, ELBACHARI, E., ENNAJAR, S., & ET-TALEBY, A. (2026). A Deep Learning Approach for Classifying Generated Geometric Pattern Datasets. Statistics, Optimization & Information Computing. https://doi.org/10.19139/soic-2310-5070-4116
License
Copyright (c) 2026 yassine AIT LAHCEN, Essaid ELBACHARI, Slimane ENNAJAR, Abdelilah ET-TALEBY
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
