Furniture Image Recognition with CBAM-Enhanced MobileNet-Based Deep Learning: A Lightweight and Efficient Approach

Öz

Automatic recognition and classification of furniture images is a difficult problem due to the high visual similarity between classes, complex shape-texture structures, and limited data diversity. In this study, we propose a lightweight, fast, and mobile-friendly deep learning framework to solve this problem. The proposed method integrates the Convolutional Block Attention Module (CBAM) into MobileNet-based architectures, aiming to both focus on critical regions and increase feature extraction capacity. Furthermore, data augmentation strategies and multi-scale representation are applied to enhance the generalization performance of the models. Experimental studies are conducted on a publicly available image dataset consisting of 32 different furniture classes. The results show that the MobileNetV1 model optimized with CBAM delivers the highest performance with an accuracy rate of 99.78%. While MobileNetV1 achieved 98.71% accuracy in its pure form, the accuracy dropped to 98.44% after data augmentation. Integrating CBAM increased the accuracy to 99.78%, but the combined use of CBAM and data augmentation yielded a relatively lower success rate of 99.57%. The MobileNetV2 model also demonstrated stable performance with CBAM and data augmentation, with 99.46% accuracy. The NASNetMobile model, on the other hand, achieved consistent success with CBAM, with 98.71% accuracy. These findings demonstrate the critical importance of model architecture, attention mechanisms, and data diversity in furniture image recognition. Furthermore, the low computational cost of the proposed approach demonstrates its suitability for real-time applications in mobile and embedded systems.

Anahtar Kelimeler

• Furniture classification
• Artificial intelligence
• Deep learning
• CNN
• Woodworking

CBAM Destekli Mobilenet Tabanlı Derin Öğrenme İle Mobilya Görüntü Tanıma: Hafif ve Etkin Bir Yaklaşım

Öz

Mobilya görüntülerinin otomatik olarak tanınması ve sınıflandırılması; sınıflar arasındaki yüksek görsel benzerlik, karmaşık şekil–doku yapıları ve veri çeşitliliğinin sınırlı olması nedeniyle zor bir problemdir. Bu çalışmada söz konusu problemi çözmek amacıyla hafif, hızlı ve mobil cihazlarda çalışmaya uygun bir derin öğrenme yapısı önerilmiştir. Önerilen yöntemde, MobileNet tabanlı mimarilere Convolutional Block Attention Module (CBAM) entegre edilerek, modelin hem kritik bölgelere odaklanması hem de özellik çıkarım kapasitesinin artırılması hedeflenmiştir. Ayrıca veri artırma stratejileri ve çok ölçekli temsil gücü, modellerin genelleme performansını güçlendirmek için uygulanmıştır. Deneysel çalışmalar, 32 farklı mobilya sınıfından oluşan kamuya açık bir görüntü veri kümesi üzerinde gerçekleştirilmiştir. Elde edilen sonuçlar, CBAM ile optimize edilen MobileNetV1 modelinin %99,78 doğruluk oranıyla en yüksek performansı sunduğunu göstermektedir. MobileNetV1 yalın hâlde %98,71 doğruluk sağlarken, veri artırma uygulaması sonrasında doğruluk %98,44’e düşmüştür. CBAM entegrasyonu doğruluğu %99,78’e yükseltmiş, ancak CBAM ve veri artırmanın birlikte kullanımı %99,57 ile nispeten daha düşük bir başarı sergilemiştir. MobileNetV2 modeli de CBAM ve veri artırma ile %99,46 doğrulukla dengeli bir performans göstermiştir. NASNetMobile modeli ise CBAM ile %98,71 doğrulukla istikrarlı bir başarı sunmuştur. Bu bulgular, model mimarisi, dikkat mekanizmaları ve veri çeşitliliğinin mobilya görüntü tanımada kritik öneme sahip olduğunu ortaya koymaktadır. Ayrıca önerilen yaklaşımın düşük hesaplama maliyeti, yöntemin mobil ve gömülü sistemlerde gerçek zamanlı uygulamalar için uygun olduğunu göstermektedir.

Anahtar Kelimeler

• Mobilya sınıflandırma
• Yapay zeka
• Derin öğrenme
• ESA
• Ağaç işleri

Kaynakça

1. Binder, A., Müller, K. R., & Kawanabe, M. (2012). On taxonomies for multi-class image categorization. International Journal of Computer Vision, 99(3), 281–301. https://doi.org/10.1007/s11263-010-0417-8
2. Chen, L., Li, S., Bai, Q., Yang, J., Jiang, S., & Miao, Y. (2021). Review of image classification algorithms based on convolutional neural networks. Remote Sensing, 13(22), 4712. https://doi.org/10.3390/rs13224712
3. Chen, W. T., Liu, W. C., & Chen, M. S. (2010). Adaptive color feature extraction based on image color distributions. IEEE Transactions on Image Processing, 19(8), 2005–2016. https://doi.org/10.1109/TIP.2010.2051753
4. Deng, L., & Yu, D. (2014). Deep learning: Methods and applications. Foundations and Trends in Signal Processing, 7(3–4), 197–387. http://dx.doi.org/10.1561/2000000039
5. Dhingra, S., & Bansal, P. (2020). Experimental analogy of different texture feature extraction techniques in image retrieval systems. Multimedia Tools and Applications, 79(37-38), 27391–27406. https://doi.org/10.1007/s11042-020-09317-3
6. Du, X. (2021, November). FISC: Furniture image style classification model based on Gram transformation. In Proceedings of the 3rd International Conference on Advanced Information Science and System (pp. 1-5). https://doi.org/10.1145/3503047.3503071
7. Duchi, J., Hazan, E., & Singer, Y. (2011). Adaptive subgradient methods for online learning and stochastic optimization. Journal of Machine Learning Research, 12(61), 2121–2159.
8. Ergün, E., & Kılıç, K. (2021). Derin öğrenme ile artırılmış görüntü seti üzerinden cilt kanseri tespiti. Black Sea Journal of Engineering and Science, 4(4), 192–200. https://doi.org/10.34248/bsengineering.938520

9. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Regularization for deep learning. In Deep learning (pp. 216–261). MIT Press.
10. Haralick, R. M., Shanmugam, K., & Dinstein, I. H. (2007). Textural features for image classification. IEEE Transactions on systems, man, and cybernetics, (6), 610-621. https://doi.org/10.1109/TSMC.1973.4309314
11. Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., & Adam, H. (2017). MobileNets: Efficient convolutional neural networks for mobile vision applications (arXiv:1704.04861). arXiv. https://doi.org/10.48550/arXiv.1704.04861
12. Howard, A., Sandler, M., Chu, G., Chen, L. C., Chen, B., Tan, M., Wang, W., Zhu, Y., Pang, R., Vasudevan, V., Le, Q. V., & Adam, H. (2019). Searching for MobileNetV3. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 1314–1324).
13. Hu, Z., Wen, Y., Liu, L., Jiang, J., Hong, R., Wang, M., & Yan, S. (2017). Visual classification of furniture styles. ACM Transactions on Intelligent Systems and Technology, 8(5), 1–20. https://doi.org/10.1145/3065951
14. Kas, M., Ruichek, Y., & Messoussi, R. (2021). New framework for person-independent facial expression recognition combining textural and shape analysis through new feature extraction approach. Information Sciences, 549, 200–220. https://doi.org/10.1016/j.ins.2020.10.065
15. Kılıç, K., Atacak, İ., & Doğru, İ. A. (2025). FABLDroid: Malware detection based on hybrid analysis with factor analysis and broad learning methods for android applications. Engineering Science and Technology, an International Journal, 62, 101945. https://doi.org/10.1016/j.jestch.2024.101945
16. Kılıç, K., Özcan, U., Kılıç, K., & Doğru, İ. (2024). Using deep learning techniques furniture image classification. Politeknik Dergisi, 27(5), 1903–1911. https://doi.org/10.2339/politeknik.1315328
17. Kingma, D. P., & Ba, J. (2014). Adam: A method for stochastic optimization (arXiv:1412.6980). arXiv. https://doi.org/10.48550/arXiv.1412.6980
18. Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017). ImageNet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84–90. https://doi.org/10.1145/3065386
19. LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. (2002). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278-2324. https://doi.org/10.1109/5.726791
20. Li, Y., Zhang, X., & Wang, Y. (2022). MFNet: Multi-level feature extraction and fusion network for point cloud classification. Remote Sensing, 14(22), 5707. https://doi.org/10.3390/rs14225707
21. Litjens, G., Kooi, T., Bejnordi, B. E., Setio, A. A. A., Ciompi, F., Ghafoorian, M., van der Laak, J. A. W. M., & van Ginneken, B. (2017). A survey on deep learning in medical image analysis. Medical Image Analysis, 42, 60–88. https://doi.org/10.1016/j.media.2017.07.005
22. Manavis, A., Minaoglou, P., Efkolidis, N., & Kyratsis, P. (2024). Digital customization for product design and manufacturing: A case study within the furniture industry. Electronics, 13(13), 2483. https://doi.org/10.3390/electronics13132483
23. Mingqiang, Y., Kidiyo, K., & Joseph, R. (2008). A survey of shape feature extraction techniques. Pattern Recognition Letters, 15(7), 43–90.
24. Nanni, L., Ghidoni, S., & Brahnam, S. (2017). Handcrafted vs. non-handcrafted features for computer vision classification. Pattern Recognition, 71, 158–172. https://doi.org/10.1016/j.patcog.2017.05.025
25. Ojala, T., Pietikäinen, M., & Mäenpää, T. (2002). Multiresolution gray-scale and rotation-invariant texture classification with local binary patterns. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(7), 971–987. https://doi.org/10.1109/TPAMI.2002.1017623
26. Pan, S. J., & Yang, Q. (2009). A survey on transfer learning. IEEE Transactions on knowledge and data engineering, 22(10), 1345-1359. https://doi.org/10.1109/TKDE.2009.191
27. Peryanto, A., Yudhana, A., & Umar, R. (2020). Klasifikasi citra menggunakan convolutional neural network dan k-fold cross validation. Journal of Applied Informatics and Computing, 4(1), 45–51. https://doi.org/10.30871/jaic.v4i1.2017
28. Pratama, R. A., Achmadi, S., & Auliasari, K. (2022). Penerapan metode convolutional neural network pada aplikasi deteksi wajah pengunjung perpustakaan. JATI (Jurnal Teknik Informatika), 6(1), 253–258. https://doi.org/10.36040/jati.v6i1.4517
29. Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., & Chen, L. C. (2018). MobileNetV2: Inverted residuals and linear bottlenecks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (pp. 4510–4520).
30. Saxen, F., Werner, P., Handrich, S., Othman, E., Dinges, L., & Al-Hamadi, A. (2019). Face attribute detection with MobileNetV2 and NASNet-Mobile. In Proceedings of the International Symposium on Image and Signal Processing and Analysis (pp. 176–180). https://doi.org/10.1109/ISPA.2019.8868585
31. Schmidhuber, J. (2015). Deep learning in neural networks: An overview. Neural Networks, 61, 85–117. https://doi.org/10.1016/j.neunet.2014.09.003
32. Therrien, A. (2024). Image classification – 32 classes – Fourniture [Data set]. Kaggle. https://www.kaggle.com/anthonytherrien/datasetsimage-classification-dataset-32-classes (Erişim tarihi: 6 Haziran 2024).
33. Wan, Z., Yang, H., Xu, J., Mu, H., & Qi, D. (2024). BAESA: Multi-scale feature fusion-based bilinear attention convolutional neural network for wood NIR classification. Journal of Forestry Research, 35(1), 4. https://doi.org/10.1007/s11676-023-01652-z
34. Xue, Y., Tong, Y., & Neri, F. (2022). An ensemble of differential evolution and Adam for training feed-forward neural networks. Information Sciences, 608, 453–471. https://doi.org/10.1016/j.ins.2022.06.036
35. Yang, X., Gao, Y., Zhang, S., Ge, Z., & Zhou, Y. (2022). Research on rosewood micro image classification method based on feature fusion and ELM. Journal of Renewable Materials, 10(12), 3587. https://doi.org/10.32604/jrm.2022.022300
36. Ye, H., Zhu, X., Liu, C., Yang, L., & Wang, A. (2022). Furniture image classification based on depthwise group over-parameterized convolution. Electronics, 11(23), 3889. https://doi.org/10.3390/electronics11233889
37. Yu, J., & Liu, J. (2022). Multiple granularities generative adversarial network for recognition of wafer map defects. IEEE Transactions on Industrial Informatics, 18(3), 1674–1683. https://doi.org/10.1109/TII.2021.3092372
38. Zhang, Y., Li, X., Chen, W., & Zang, Y. (2024). Image classification based on low-level feature enhancement and attention mechanism. Neural Processing Letters, 56(4), 217. https://doi.org/10.1007/s11063-024-11680-3