Deep learning-based method for array ultrasonic total focus imaging of internal cracks in RC beams

Abstract

Implementing the conventional total focus method (TFM) for visualizing internal damage in reinforced concrete (RC) is beset with computational challenges and a high dependence on physical principles. To overcome these challenges, an efficient total focus imaging method based on deep learning is proposed. This method deals with array ultrasonic time-domain signals from cracked RC beams. A deep neural network (DNN) employing a feature extraction + multilevel feature fusion + matrix construction architecture was developed; this architecture enabled the DNN to learn the underlying physical principles of the TFM. The architecture effectively transformed ultrasonic time-domain signals into a B-scan matrix. Training, validation, and testing data were collected by measuring eight RC beams with preset artificial cracks using a low-frequency shear wave array ultrasonic system. The results demonstrated that the reconstructed B-scan matrices had a peak signal-to-noise ratio of 26.94 dB and a structural similarity index of 0.978. Furthermore, the proposed method required 42% fewer floating-point operations compared with physics-based calculations, achieving total focus imaging with lower computational cost. The study facilitates the advancement of ultrasonic total focus imaging of RC structures from physics-based methods to data-driven methods without requiring prior physical knowledge, thereby providing robust support for further nondestructive evaluation and quantitative analysis.

Keywords

total focus method / array ultrasound imaging / reinforced concrete beam / time-domain signal / deep learning

References

[1] HOEGH K, KHAZANOVICH L, MASER K, et al. Evaluation of ultrasonic technique for detecting delamination in asphalt pavements[J]. Transportation Research Record: Journal of the Transportation Research Board, 2012, 2306(1): 105-110.

[2] GHOSH D, KUMAR R, GANGULI A, et al. Nondestructive evaluation of rebar corrosion-induced damage in concrete through ultrasonic imaging[J]. Journal of Materials in Civil Engineering, 2020, 32(10): 04020294.

[3] DE LA HAZA A O, SAMOKRUTOV A A, SAMOKRUTOV P A. Assessment of concrete structures using the Mira and Eyecon ultrasonic shear wave devices and the SAFT-C image reconstruction technique[J]. Construction and Building Materials, 2013, 38: 1276-1291.

[4] KWON H, JOH C, CHIN W J. 3D internal visualization of concrete structure using multifaceted data for ultrasonic array pulse-echo tomography[J]. Sensors, 2021, 21(19): 6681.

[5] JING H C. Improving SAFT imaging technology for ultrasonic detection of concrete structures[J]. Journal of Applied Sciences, 2013, 13(21): 4363-4370.

[6] FENG K, ZHAO Q, QIU Y P. Damage imaging in mesoscale concrete modeling based on the ultrasonic time-reversal technique[J]. Acta Mechanica Solida Sinica, 2020, 33(1): 61-70.

[7] LIU H, QI Y, CHEN Z J, et al. Ultrasonic inspection of grouted splice sleeves in precast concrete structures using elastic reverse time migration method[J]. Mechanical Systems and Signal Processing, 2021, 148: 107152.

[8] LIU H, XIA H Y, ZHUANG M W, et al. Reverse time migration of acoustic waves for imaging based defects detection for concrete and CFST structures[J]. Mechanical Systems and Signal Processing, 2019, 117: 210-220.

[9] SCHICKERT M, KRAUSE M, MÜLLER W. Ultrasonic imaging of concrete elements using reconstruction by synthetic aperture focusing technique[J]. Journal of Materials in Civil Engineering, 2003, 15(3): 235-246.

[10] GE L L, LI Q S, WANG Z G, et al. High-resolution ultrasonic imaging technology for the damage of concrete structures based on total focusing method[J]. Computers and Electrical Engineering, 2023, 105: 108526.

[11] HOLMES C, DRINKWATER B W, WILCOX P D. Post-processing of the full matrix of ultrasonic transmit-receive array data for non-destructive evaluation[J]. NDT & E International, 2005, 38(8): 701-711.

[12] KUCHIPUDI S T, GHOSH D. An ultrasonic wave- based framework for imaging internal cracks in concrete[J]. Structural Control and Health Monitoring, 2022, 29(12): e3108.

[13] LI F Z, LUO Y, ZHANG B. Research on ultrasonic damage imaging based on total focus method[C]//2019 14th Symposium on Piezoelectrcity, Acoustic Waves and Device Applications (SPAWDA). Shijiazhuang, China, 2019: 1-7.

[14] GU X F, ZHANG Q Y, SUN X Y, et al. Magnet array-actuated steerable flexible robot with beacon-TFM ultrasonic position sensing for robotic neurosurgery[C]// 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). Detroit, MI, USA, 2023: 10275-10280.

[15] NANEKAR P, KUMAR A, JAYAKUMAR T. SAFT-assisted sound beam focusing using phased arrays (PA-SAFT) for non-destructive evaluation[J]. Nondestructive Testing and Evaluation, 2015, 30(2): 105-123.

[16] GUO X F, HAN Y, NIE P F. Ultrasound imaging algorithm: Half-matrix focusing method based on reciprocity[J]. Mathematical Problems in Engineering, 2021, 2021(1): 8888469.

[17] PIEDADE L P, PAINCHAUD-APRIL G, LE DUFF A, et al. Compressive sensing strategy on sparse array to accelerate ultrasonic TFM imaging[J]. IEEE Transactions on Ultrasonics, Ferroelectrics, and Frequency Control, 2023, 70(6): 538-550.

[18] ZHANG C W, CHUN Q, MA Y K, et al. Research on damage detection of ancient stone arch bridges based on spatio-temporal difference graph convolutional neural network[J]. Journal of Southeast University (Natural Science Edition), 2025, 55(2):370-379. (in Chinese)

[19] WANG S, ZHANG L J, YIN G J. Defect identification method for steel surfaces based on improved YOLOv5[J]. Journal of Southeast University (English Edition), 2024, 40 (1): 49-57.

[20] LI D, CHEN Q F, WANG H, et al. Deep learning- based acoustic emission data clustering for crack evaluation of welded joints in field bridges[J]. Automation in Construction, 2024, 165: 105540.

[21] LI D, WANG Y, YAN W J, et al. Acoustic emission wave classification for rail crack monitoring based on synchrosqueezed wavelet transform and multi-branch convolutional neural network[J]. Structural Health Monitoring, 2021, 20(4): 1563-1582.

[22] LUIJTEN B, COHEN R, DE BRUIJN F J, et al. Deep learning for fast adaptive beamforming[C]// ICASSP 2019—2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Brighton, UK, 2019: 1333-1337.

[23] JEDRUSIAK M D, WEICHERT F. A deep learning approach for denoising air-coupled ultrasonic responds data[J]. International Journal of Artificial Intelligence & Applications, 2020, 11(4): 15-28.

[24] MUNIR N, PARK J, KIM H J, et al. Performance enhancement of convolutional neural network for ultrasonic flaw classification by adopting autoencoder[J]. NDT & E International, 2020, 111: 102218.

[25] SONG S P, JIA H, CHEN D. Ultrasonic image artifact removal technique based on SoS kernel[J]. Journal of Southeast University (English Edition), 2024, 40(1): 80-88.

Funding

Science & Technology Specific Project of Jiangsu Province(BZ2024047)

Key R&D Program of Ningbo(2024H013)

National Natural Science Foundation of China(W2412092)