Journal of Engineering, Project, and Production Management, 2026, 16(6), 2025-358
A Library Retrieval Model Using Feature Fusion and Cross-Modal Retrieval Techniques
1 Librarian,
Information Consultation Department, Dezhou University, Dezhou, 253023,
China, E-mail: dzxylzc@163.com (corresponding author).
2 Associate Research Librarian,
Discipline Inspection Commission, Dezhou University, Dezhou, 253023,
China, E-mail: wangfang0058@163.com
Project Management
Received December 25, 2025; revised February 10, 2026; August 7, 2026; accepted September 8, 2026
Available online September 20, 2026
Abstract: This study proposes a library retrieval model that uses feature fusion and cross-modal retrieval techniques to solve the weak processing of multi-modal resources and the low retrieval efficiency of current library retrieval methods. The model first uses deep learning algorithms to extract features from multi-modal library resources. It then fuses extracted image features and text features. After that, it builds a similarity matrix that measures the relation between images and texts. The model uses this matrix to perform a hash function search and completes efficient image and text retrieval in the library. Feature extraction experiments show that the model achieves the highest mean average precision in cross-modal retrieval scenarios involving textual queries and visual inputs, as well as visual queries and textual outputs, reaching 0.8953 and 0.9034, respectively. The model also obtains the highest accuracy under various noise conditions, with a mean accuracy of 95.08%. These findings demonstrate the proposed model's ability to enhance the efficiency of library resource retrieval and facilitate more effective resource management and utilization.
Keywords: Feature fusion; cross-modal retrieval techniques; library retrieval model; deep learning; hash function.
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: Li, Z. and Wang, F. (2026). A Library Retrieval Model Using Feature Fusion and Cross-Modal Retrieval Techniques. Journal of Engineering, Project, and Production Management, 16(6), 2025-358.
