Journal of Engineering, Project, and Production Management, 2026, 16(6), 2026-224
Intelligent Ancient Pottery Porcelain Type Features Recognition
Lecturer, Publicity and United Front
Department, Zibo Polytechnic University, Zibo, 55300, China, E-mail:
mayuan88662025@163.com
Project Management
Received February 10, 2026; revised March 31, 2026; accepted September 10, 2026
Available online September 20, 2026
Abstract: As an important carrier of historical culture and craft technology, accurate identification of ancient pottery and porcelain shapes is crucial for cultural relic dating, archaeological research, and cultural heritage preservation. At present, the identification of ancient ceramic types relies mainly on traditional manual methods, which are subjective, inefficient, and limited in their ability to detect subtle morphological features. To address these limitations, this study proposes a Multidimensional Data Representation (MDR) and an intelligent recognition method for ancient ceramic shape features based on Artificial Intelligence (AI). A multimodal data fusion framework, "image-3D-semantics," is constructed to achieve comprehensive and fine-grained characterization of ceramic shape features, and an improved Deep Support Vector Machine (Deep-SVM) is adopted for intelligent recognition. Experiments are conducted on a self-collected dataset of 900 ancient ceramic samples covering 7 categories (bowls, plates, bottles, cans, kettles, stoves, cups/pots), with a train-test split ratio of 7:3. Baseline comparisons include traditional manual features + SVM, single image feature + SVM, single 3D feature + SVM, and basic CNN features + SVM. On this basis, an intelligent recognition technique is developed to improve the accuracy and efficiency of ancient ceramic identification. Results show that after integrating multi-source information, the proposed model increases feature coverage to 92%, discrimination to 88%, reduces data redundancy to 12%, and improves the fit to real-world models to 90%. Overall, the model demonstrates clear advantages in comprehensiveness, differentiation, efficiency, and authenticity. The findings provide new technical means and methodological insights for ancient ceramic research, supporting its digital and intelligent development.
Keywords: Artificial intelligence (AI) technology; ancient pottery porcelain type; intelligent recognition; multidimensional data; feature representation.
Copyright © Journal of Engineering, Project, and Production Management (EPPM-Journal).
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License.
Requests for reprints and permissions at eppm.journal@gmail.com.
Citation: Ma, Y. (2026). Intelligent Ancient Pottery Porcelain Type Features Recognition. Journal of Engineering, Project, and Production Management, 16(6), 2026-224.
