Improved Lightweight Deep Learning Model for the Fine-Feature Classification of Pipe Components


  • Changxing Chen School of Computer Science and Engineering, Taylor’s University, Subang Jaya,Malaysia.
  • Afizan Azman School of Computer Science and Engineering, Taylor’s University, Subang Jaya,Malaysia.


Object classification, Subtle image features, Lightweight network, Industrial


In the field of industrial manufacturing, the most common application of machine vision is object classification. Traditional models face challenges in classifying objects with small features and have yet to develop a complete universal algorithm, resulting in lower accuracy. To address these issues and enable the model's use on portable embedded devices, as well as facilitate the classification of pipe components with subtle internal features, a lightweight deep learning model has been developed. This model utilizes a relatively small dataset comprising specific types of work pieces occurring in actual factory production. The dataset involves the use of subtle image features for the classification of pipe components. The proposed model combines a fine-tuning module to capture multi-scale features of the input image and utilizes attention mechanisms to enhance the model's generalization ability. Not only enhances detection accuracy but also achieves network lightweight, with an accuracy of 96.33%. A comparison with other models demonstrates an improvement in accuracy of at least 4%, along with a significant reduction in both total and training parameters, meeting usability requirements, possessing lower computational complexity is essential to ensure fast and efficient operation in scenarios with embedded devices, mobile devices, or other resource-constrained environments.


Download data is not yet available.


S. G. Karnaukh, O. E. Markov, A. A. Shapoval, and N. S. Hrudkina, "Selecting a cutting method for workpieces before stamping using synergetic fracture criteria and a deformability limit determination technique for separating processes," International Journal of Advanced Manufacturing Technology, Article vol. 129, no. 11-12, pp. 5447-5455, 2023, doi: 10.1007/s00170-023-12627-z.

G. Zhou et al., "A new algorithm for chatter quantification and milling instability classification based on surface analysis," Mechanical Systems and Signal Processing, Article vol. 204, 2023, Art no. 110816, doi: 10.1016/j.ymssp.2023.110816.

F. Khan, K. Kamal, T. A. H. Ratlamwala, M. Alkahtani, M. Almatani, and S. Mathavan, "Tool Health Classification in Metallic Milling Process Using Acoustic Emission and Long Short-Term Memory Networks: A Deep Learning Approach," IEEE Access, Article vol. 11, pp. 126611-126633, 2023, doi: 10.1109/ACCESS.2023.3328582.

P. Zhang et al., "Improving generalisation and accuracy of on-line milling chatter detection via a novel hybrid deep convolutional neural network," Mechanical Systems and Signal Processing, Article vol. 193, 2023, Art no. 110241, doi: 10.1016/j.ymssp.2023.110241.

S. Zhu, G. Fu, Y. Zheng, Z. Li, and J. Yang, "Universal Surface Texture Modeling Method for Five-axis Surface Milling," Zhongguo Jixie Gongcheng/China Mechanical Engineering, Article vol. 34, no. 16, pp. 1946-1957, 2023, doi: 10.3969/j.issn.1004-132X.2023.16.008.

A. Fertig, M. Weigold, and Y. Chen, "Machine Learning based quality prediction for milling processes using internal machine tool data," Advances in Industrial and Manufacturing Engineering, Article vol. 4, 2022, Art no. 100074, doi: 10.1016/j.aime.2022.100074.

D. A. Molitor, C. Kubik, R. H. Hetfleisch, and P. Groche, "Workpiece image-based tool wear classification in blanking processes using deep convolutional neural networks," Production Engineering, Article vol. 16, no. 4, pp. 481-492, 2022, doi: 10.1007/s11740-022-01113-2.

C. Chen, A. Abdullah, S. H. Kok, and D. T. K. Tien, "Review of Industry Workpiece Classification and Defect Detection using Deep Learning," International Journal of Advanced Computer Science and Applications, Article vol. 13, no. 4, pp. 329-340, 2022, doi: 10.14569/IJACSA.2022.0130439.

Q. Li, Z. Luo, H. Chen, and C. Li, "An Overview of Deeply Optimized Convolutional Neural Networks and Research in Surface Defect Classification of Workpieces," IEEE Access, Article vol. 10, pp. 26443-26462, 2022, doi: 10.1109/ACCESS.2022.3157293.

C. H. Wang et al., "Deep Learning-based Diagnosis and Localization of Pneumothorax on Portable Supine Chest X-ray in Intensive and Emergency Medicine: A Retrospective Study," Journal of medical systems, Article vol. 48, no. 1, p. 1, 2023, doi: 10.1007/s10916-023-02023-1.

D. Wu, J. Yang, M. U. Ahsan, and K. Wang, "Classification of integers based on residue classes via modern deep learning algorithms," Patterns, Article vol. 4, no. 12, 2023, Art no. 100860, doi: 10.1016/j.patter.2023.100860.

Y. Dou, J. Xia, M. Fu, Y. Cai, X. Meng, and Y. Zhan, "Identification of epileptic networks with graph convolutional network incorporating oscillatory activities and evoked synaptic responses," NeuroImage, Article vol. 284, 2023, Art no. 120439, doi: 10.1016/j.neuroimage.2023.120439.

E. Tjoa, H. J. Khok, T. Chouhan, and C. Guan, "Enhancing the confidence of deep learning classifiers via interpretable saliency maps," Neurocomputing, Article vol. 562, 2023, Art no. 126825, doi: 10.1016/j.neucom.2023.126825.

Z. H. Ren et al., "DeepMPF: deep learning framework for predicting drug–target interactions based on multi-modal representation with meta-path semantic analysis," Journal of Translational Medicine, Article vol. 21, no. 1, 2023, Art no. 48, doi: 10.1186/s12967-023-03876-3.

D. Tabernik, M. Šuc, and D. Skočaj, "Automated detection and segmentation of cracks in concrete surfaces using joined segmentation and classification deep neural network," Construction and Building Materials, Article vol. 408, 2023, Art no. 133582, doi: 10.1016/j.conbuildmat.2023.133582.

X. Xiahou, Z. Li, J. Xia, Z. Zhou, and Q. Li, "A Feature-Level Fusion-Based Multimodal Analysis of Recognition and Classification of Awkward Working Postures in Construction," Journal of Construction Engineering and Management, Article vol. 149, no. 12, 2023, Art no. 04023138, doi: 10.1061/JCEMD4.COENG-13795.

A. N. Arun, P. Maheswaravenkatesh, and T. Jayasankar, "Facial Micro Emotion Detection and Classification Using Swarm Intelligence based Modified Convolutional Network," Expert Systems with Applications, Article vol. 233, 2023, Art no. 120947, doi: 10.1016/j.eswa.2023.120947.

Y. Cai, Z. Yao, X. Cheng, Y. He, S. Li, and J. Pan, "Deep metric learning framework combined with Gramian angular difference field image generation for Raman spectra classification based on a handheld Raman spectrometer," Spectrochimica Acta - Part A: Molecular and Biomolecular Spectroscopy, Article vol. 303, 2023, Art no. 123085, doi: 10.1016/j.saa.2023.123085.

O. F. Abd-Elaziz, M. Abdalla, and R. A. Elsayed, "Deep Learning-Based Automatic Modulation Classification Using Robust CNN Architecture for Cognitive Radio Networks," Sensors, Article vol. 23, no. 23, 2023, Art no. 9467, doi: 10.3390/s23239467.

R. Yousaf et al., "Satellite Imagery-Based Cloud Classification Using Deep Learning," Remote Sensing, Article vol. 15, no. 23, 2023, Art no. 5597, doi: 10.3390/rs15235597.

T. L. Huang et al., "Transfer learning with CNNs for efficient prostate cancer and BPH detection in transrectal ultrasound images," Scientific Reports, Article vol. 13, no. 1, 2023, Art no. 21849, doi: 10.1038/s41598-023-49159-1.

Y. Chai, H. Yu, L. Xu, D. Li, and Y. Chen, "Deep Learning Algorithms for Sonar Imagery Analysis and Its Application in Aquaculture: A Review," IEEE Sensors Journal, Article vol. 23, no. 23, pp. 28549-28563, 2023, doi: 10.1109/JSEN.2023.3324438.

H. Liu et al., "A Deep Learning Neural Network Method Using Linear Eigenvalue Statistics for Schizophrenic EEG Data Classification," Mathematics, Article vol. 11, no. 23, 2023, Art no. 4776, doi: 10.3390/math11234776.




How to Cite

Chen, C. ., & Azman, A. . (2024). Improved Lightweight Deep Learning Model for the Fine-Feature Classification of Pipe Components. International Journal of Intelligent Systems and Applications in Engineering, 12(15s), 520–529. Retrieved from



Research Article