RADIOELECTRONIC AND COMPUTER SYSTEMS

Models and training methods for water-level classification analysis on the footage of sewage pipe inspections have been developed and investigated. The object of the research is the process of water-level recognition, considering the spatial and temporal context during the inspection of sewage pipes. The subject of the research is a model and machine learning method for water-level classification analysis on video sequences of pipe inspections under conditions of limited size and an unbalanced set of training data. A four-stage algorithm for training the classifier is proposed. At the first stage of training, training occurs with a softmax triplet loss function and a regularizing component to penalize the rounding error of the network output to a binary code. The next step is to define a binary code (reference vector) for each class according to the principles of error-correcting output codes, but considering the intraclass and interclass relations. The computed reference vector of each class is used as the target label of the sample for further training using the joint cross-entropy loss function. The last stage of machine learning involves optimizing the parameters of the decision rules based on the information criterion to account for the boundaries of deviation of the binary representation of the observations of each class from the corresponding reference vectors. As a classifier model, a combination of 2D convolutional feature extractor for each frame and temporal network to analyze inter-frame dependencies is considered. The different variants of the temporal network are compared. We consider a 1D regular convolutional network with dilated convolutions, 1D causal convolutional network with dilated convolutions, recurrent LSTM-network, recurrent GRU-network. The performance of the models is compared by the micro-averaged metric F1 computed on the test subset. The results obtained on the dataset from Ace Pipe Cleaning (Kansas City, USA) confirm the suitability of the model and training method for practical use, the obtained value of F1-metric is 0.88. The results of training by the proposed method were compared with the results obtained using the traditional method. It was shown that the proposed method provides a 9 % increase in the value of micro-averaged F1-measure.

sewer pipe; inspection; classification analysis; convolutional neural network; recurrent neural network; loss function; regularization; information-extreme machine learning

Moradi, S., Zayed, T., Golkhoo, F. Review On Computer Aided Sewer Pipeline Defect Detection And Condition Assessment. Infrastructures, 2019, vol. 4, no. 1, Article Id: 10. DOI: 10.3390/infrastructures4010010.

Haurum, J. B., Moeslund, T. B. A Survey on image-based automation of CCTV and SSET sewer inspections. Automation in Construction, 2020, vol. 111, Article Id: 10306. DOI: 10.1016/j.autcon.2019.103061.

American Society of Civil Engineers (ASCE). 2017 Infrastructure Report Card – Wastewater. Available at: https://www.infrastructurereportcard.org/wp-content/uploads/2017/01/Wastewater-Final.pdf (accessed 05.05.2021).

Cheng, G. C. P. Wang M. Automated detection of sewer pipe defects in closed-circuit television images using deep learning techniques. Automation in Construction, 2018,vol. 95, pp. 155-171. DOI: 10.1016/j.autcon.2018.08.006.

Lim, B., Zohren, S. Time-series forecasting with deep learning: a survey. Philosofical Transactions of the Royal Society A, 2021, vol. 379, no. 2194. 14 p. DOI: 10.1098/rsta.2020.0209.

Kirstein, S., Müller, K., Walecki-Mingers, M., Deserno, T. M. Robust adaptive flow line detection in sewer pipes. Automation in Construction, 2012, vol. 21, pp. 24-31. DOI: 10.1016/j.autcon.2011.05.009.

Halfawy, M. R., Hengmeechai, J. Integrated Vision-Based System for Automated Defect Detection in Sewer Closed Circuit Television Inspection Videos. Journal of Computing in Civil Engineering, 2015, vol. 29, no. 1. DOI: 10.1061/(ASCE)CP.1943-5487.0000312.

Ji, H. W., Yoo, S. S., Lee, B.-J., Koo, D. D., Kang, J.-H. Measurement of Wastewater Discharge in Sewer Pipes Using Image Analysis. Water, 2020, vol. 12, no. 6, Article Id: 1771. DOI: 10.3390/w12061771.

Zhan, H., Shi, B., Duan, L.-Y., Kot, A. C. Deepshoe: An Improved Multi-Task View-Invariant CNN For Street-To-Shop Shoe Retrieval. Computer Vision and Image Understanding, 2019, vol. 180, pp. 23-33. DOI: 10.1016/j.cviu.2019.01.001.

Zhang, B., Tondi, B., Lv, X., Barni, M. Challenging the Adversarial Robustness of DNNs Based on Error-Correcting Output Codes. Security and Communication Networks, 2020, vol. 2020, Article Id: 8882494. DOI: 10.1155/2020/8882494.

Dovbysh, A. S., Shelekhov, I. V., Khibovs'ka, Yu. O., Matyash, O. V. Informatsiyno-analitychna systema otsinyuvannya vidpovidnosti suchasnym vymoham navchal'noho kontentu spetsial'nosti kiberbezpeka [Information and analytical system for assessing the compliance of educational content specialties ciber security with modern requirements]. Radioelektronni i komp'uterni sistemi – Radioelectronic and computer systems, 2021, vol. 1, pp. 70-80. DOI: 10.32620/reks.2021.1.06.