Toward Amodal Instance Segmentation of Overlapping Concrete Aggregates via Occlusion-Conditioned Residual Correction

I Gede Teguh Satya Dharma, I Nyoman Gede Arya Astawa, I Nyoman Suardika, Made Pradnyana Ambara
Pages 2765-2780
Abstract Views 41
PDF Downloads 19
Crossref Citations -
DOI https://doi.org/10.47738/jads.v7i4.1569

Abstract


Concrete aggregate characterization is important for quality control, particularly in estimating particle size and shape distributions. Although computer vision provides a faster alternative to conventional sieve analysis, overlapping particles can hide parts of their boundaries, causing visible-only segmentation to underestimate the complete particle geometry. This study proposes an occlusion-conditioned residual correction mechanism for amodal instance segmentation of overlapping concrete aggregates. Rather than predicting the complete mask from scratch, the method preserves the observed visible region and estimates only the missing geometry along its boundary, using the visible mask as an explicit physical constraint. Local boundary information is incorporated through a ring context to guide residual prediction. The proposed mechanism is implemented within a standard Mask R-CNN framework and evaluated on a synthetic aggregate dataset with controlled occlusion levels. Compared with the corresponding direct amodal baseline, the proposed method improves AP from 48.02% to 52.69%, while ablation results show an additional gain from the ring context. Qualitative results further indicate more faithful reconstruction of irregular contours, narrow gaps, and sharp corners with less excessive mask expansion. These findings support constrained residual reconstruction as an effective mechanism for amodal aggregate segmentation under occlusion.

Keywords


Concrete Aggregates; Amodal Instance Segmentation; Occlusion Handling; Residual Correction; Mask R-CNN; Computer Vision

Full Text:

PDF

References


B. Wang, L. Yan, Q. Fu, and B. Kasal, “A Comprehensive Review on Recycled Aggregate and Recycled Aggregate Concrete,” Resour. Conserv. Recycl., vol. 171, p. 105565, 2021, doi: 10.1016/j.resconrec.2021.105565.

Z. Abedin Khan, U. Balunaini, S. Costa, and N. H. T. Nguyen, “A review on sustainable use of recycled construction and demolition waste aggregates in pavement base and subbase layers,” Cleaner Materials, vol. 13, p. 100266, 2024, doi: 10.1016/j.clema.2024.100266.

M. Coenen, D. Beyer, C. Heipke, and M. Haist, “Learning to Sieve: Prediction of Grading Curves from Images of Concrete Aggregate,” ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, vol. V-2–2022, pp. 227–235, 2022, doi: 10.5194/isprs-annals-V-2-2022-227-2022.

G. Cao et al., “Influence of aggregate shape on the flow properties of fresh concrete,” Powder Technol., vol. 415, p. 118186, 2023, doi: 10.1016/j.powtec.2022.118186.

V. M. Nguyen, T. T. Nguyen, H. M. Gioi, T. Van Phan, and R. Siddique, “Optimizing aggregate grading and residual mortar coefficient in low-cement concrete by using crushed coal mine waste rock as base layers for low-traffic roads,” Constr. Build. Mater., vol. 449, p. 138468, 2024, doi: 10.1016/j.conbuildmat.2024.138468.

L. Wang, Y. Yao, J. Li, Y. Tao, and K. Liu, “Review of Visualization Technique and Its Application of Road Aggregates Based on Morphological Features,” Applied Sciences, vol. 12, no. 20, p. 10571, 2022, doi: 10.3390/app122010571.

X. Hu, H. Fang, J. Yang, L. Fan, W. Lin, and J. Li, “Online measurement and segmentation algorithm of coarse aggregate based on deep learning and experimental comparison,” Constr. Build. Mater., vol. 327, p. 127033, 2022, doi: 10.1016/j.conbuildmat.2022.127033.

M. Moaveni, S. Wang, J. M. Hart, E. Tutumluer, and N. Ahuja, “Evaluation of Aggregate Size and Shape by Means of Segmentation Techniques and Aggregate Image Processing Algorithms,” Transportation Research Record: Journal of the Transportation Research Board, vol. 2335, no. 1, pp. 50–59, 2013, doi: 10.3141/2335-06.

W. Wang, C. Su, and H. Zhang, “Automatic segmentation of concrete aggregate using convolutional neural network,” Autom. Constr., vol. 134, p. 104106, 2022, doi: 10.1016/j.autcon.2021.104106.

Y. Xiao et al., “A novel method for predicting coarse aggregate particle size distribution based on segment anything model and machine learning,” Constr. Build. Mater., vol. 429, p. 136429, 2024, doi: 10.1016/j.conbuildmat.2024.136429.

H. Fan et al., “Rockfill material segmentation and gradation calculation based on deep learning,” Case Studies in Construction Materials, vol. 17, p. e01216, 2022, doi: 10.1016/j.cscm.2022.e01216.

J. Zhao, D. Li, and Y. Yu, “Identification of Rock Fragments after Blasting by Using Deep Learning-Based Segment Anything Model,” Minerals, vol. 14, no. 7, p. 654, 2024, doi: 10.3390/min14070654.

R. Kanke and M. Takahashi, “Amodal Instance Segmentation of Thin Objects with Large Overlaps by Seed-to-Mask Extending,” IEICE Trans. Inf. Syst., vol. E107.D, no. 7, pp. 908–911, 2024, doi: 10.1587/transinf.2023EDL8068.

M. Tran, W. Bounsavy, K. Vo, A. Nguyen, T. Nguyen, and N. Le, “ShapeFormer: Shape Prior Visible-to-Amodal Transformer-based Amodal Instance Segmentation,” in 2024 International Joint Conference on Neural Networks (IJCNN), IEEE, 2024, pp. 1–8. doi: 10.1109/IJCNN60899.2024.10650837.

R. Mohan and A. Valada, “Perceiving the Invisible: Proposal-Free Amodal Panoptic Segmentation,” IEEE Robot. Autom. Lett., vol. 7, no. 4, pp. 9302–9309, 2022, doi: 10.1109/LRA.2022.3189425.

L. Qi, L. Jiang, S. Liu, X. Shen, and J. Jia, “Amodal Instance Segmentation With KINS Dataset,” in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2019, pp. 3009–3018. doi: 10.1109/CVPR.2019.00313.

Y. Hu, J. Wang, X. Wang, and T. Guan, “Occlusion-aware particle size distribution detection of gravel material based on the improved Bilayer Convolutional Network,” Constr. Build. Mater., vol. 404, p. 133263, 2023, doi: 10.1016/j.conbuildmat.2023.133263.

Y. Hu, J. Wang, X. Wang, J. Yu, and J. Zhang, “Efficient virtual-to-real dataset synthesis for amodal instance segmentation of occlusion-aware rockfill material gradation detection,” Expert Syst. Appl., vol. 238, p. 122046, 2024, doi: 10.1016/j.eswa.2023.122046.

K. He, G. Gkioxari, P. Dollar, and R. Girshick, “Mask R-CNN,” in 2017 IEEE International Conference on Computer Vision (ICCV), IEEE, 2017, pp. 2980–2988. doi: 10.1109/ICCV.2017.322.

L. Ke, Y.-W. Tai, and C.-K. Tang, “Deep Occlusion-Aware Instance Segmentation with Overlapping BiLayers,” in 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2021, pp. 4018–4027. doi: 10.1109/CVPR46437.2021.00401.

K. Li and J. Malik, “Amodal Instance Segmentation,” in Computer Vision – ECCV 2016, vol. 9906, B. Leibe, J. Matas, N. Sebe, and M. Welling, Eds., Cham: Springer International Publishing, 2016, pp. 677–693. doi: 10.1007/978-3-319-46475-6_42.

Y. Zhu, Y. Tian, D. Metaxas, and P. Dollar, “Semantic Amodal Segmentation,” in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2017, pp. 3001–3009. doi: 10.1109/CVPR.2017.320.

K. Ehsani, R. Mottaghi, and A. Farhadi, “SeGAN: Segmenting and Generating the Invisible,” in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2018, pp. 6144–6153. doi: 10.1109/CVPR.2018.00643.

E. Ozguroglu et al., “pix2gestalt: Amodal Segmentation by Synthesizing Wholes,” in 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2024, pp. 3931–3940. doi: 10.1109/CVPR52733.2024.00377.

N. Carion, F. Massa, G. Synnaeve, N. Usunier, A. Kirillov, and S. Zagoruyko, “End-to-End Object Detection with Transformers,” in Computer Vision – ECCV 2020, vol. 12346, A. Vedaldi, H. Bischof, T. Brox, and J.-M. Frahm, Eds., Cham: Springer International Publishing, 2020, pp. 213–229. doi: 10.1007/978-3-030-58452-8_13.

B. Cheng, I. Misra, A. G. Schwing, A. Kirillov, and R. Girdhar, “Masked-attention Mask Transformer for Universal Image Segmentation,” in 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2022, pp. 1280–1289. doi: 10.1109/CVPR52688.2022.00135.

Z. Wang, H. Zheng, P. He, W. Chen, and M. Zhou, “Diffusion-GAN: Training GANs with Diffusion,” 2022, arXiv. doi: 10.48550/ARXIV.2206.02262.

S. Ren, K. He, R. Girshick, and J. Sun, “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 39, no. 6, pp. 1137–1149, 2017, doi: 10.1109/TPAMI.2016.2577031.

Z. Li, W. Ye, T. Jiang, and T. Huang, “2D Amodal Instance Segmentation Guided by 3D Shape Prior,” in Computer Vision – ECCV 2022, vol. 13689, S. Avidan, G. Brostow, M. Cissé, G. M. Farinella, and T. Hassner, Eds., Cham: Springer Nature Switzerland, 2022, pp. 165–181. doi: 10.1007/978-3-031-19818-2_10.

G. Zhan, C. Zheng, W. Xie, and A. Zisserman, “Amodal Ground Truth and Completion in the Wild,” in 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2024, pp. 28003–28013. doi: 10.1109/CVPR52733.2024.02645.

S. Raghavendra, D. Rao, S. K. Abhilash, V. M. Nookala, and P. G. Bharathi, “Elevating Amodal Segmentation Using ASH-Net Architecture for Accurate Object Boundary Estimation,” IEEE Access, vol. 11, pp. 83377–83389, 2023, doi: 10.1109/ACCESS.2023.3301724.

J. Gao et al., “Coarse-to-Fine Amodal Segmentation with Shape Prior,” in 2023 IEEE/CVF International Conference on Computer Vision (ICCV), IEEE, 2023, pp. 1262–1271. doi: 10.1109/ICCV51070.2023.00122.

J. Wei et al., “Mixture-of-Shape-Experts (MoSE): End-to-End Shape Dictionary Framework to Prompt SAM for Generalizable Medical Segmentation,” in 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), IEEE, 2025, pp. 6450–6460. doi: 10.1109/CVPRW67362.2025.00642.

X. Zhao, J. Xie, S. Chen, and J. Liu, “Convex Combination Star Shape Prior for Data-driven Image Semantic Segmentation,” in 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2025, pp. 14068–14077. doi: 10.1109/CVPR52734.2025.01313.




DOI: https://doi.org/10.47738/jads.v7i4.1569

Refbacks

  • There are currently no refbacks.


Barcode

Journal of Applied Data Sciences

ISSN:2723-6471 (Online)
Publisher:Bright Publisher
Website:http://bright-journal.org/JADS
Email:taqwa@amikompurwokerto.ac.id (principal contact)
  support@bright-journal.org (technical issues)

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0