TY - JOUR
T1 - PIPE-CovNet+ : a hyper-dense CNN for improved pipe abnormality detection
AU - Wang, Xing
AU - Thiyagarajan, Karthick
AU - Kodagoda, Sarath
AU - Sharma, Charu
PY - 2024/4/1
Y1 - 2024/4/1
N2 - In this letter, we propose a PIPE-CovNet+ model that is based on convolutional neural networks (CNNs) with multiple kernel sizes, gradient boosting techniques, and hyper- densely connected layers for abnormality detection in wastewater pipe infrastructure. The PIPE-CovNet+ model achieved 85% accuracy and 84% F1-score following assessment utilizing an open-source dataset, overcoming the constraints of imbalanced data and over-fitting issues. Furthermore, it showed greater efficacy when compared against other relevant models, displaying 3% higher accuracy and F1-score than our prior work under ideal test settings. Installing PIPE-CovNet+ into a robot system can alleviate the limitations of being tedious, laborious, and prone to human error in closed-circuit television (CCTV) inspections.
AB - In this letter, we propose a PIPE-CovNet+ model that is based on convolutional neural networks (CNNs) with multiple kernel sizes, gradient boosting techniques, and hyper- densely connected layers for abnormality detection in wastewater pipe infrastructure. The PIPE-CovNet+ model achieved 85% accuracy and 84% F1-score following assessment utilizing an open-source dataset, overcoming the constraints of imbalanced data and over-fitting issues. Furthermore, it showed greater efficacy when compared against other relevant models, displaying 3% higher accuracy and F1-score than our prior work under ideal test settings. Installing PIPE-CovNet+ into a robot system can alleviate the limitations of being tedious, laborious, and prone to human error in closed-circuit television (CCTV) inspections.
UR - https://hdl.handle.net/1959.7/uws:77723
U2 - 10.1109/LSENS.2024.3374814
DO - 10.1109/LSENS.2024.3374814
M3 - Article
SN - 2475-1472
VL - 8
JO - IEEE Sensors Letters
JF - IEEE Sensors Letters
IS - 4
M1 - 6003504
ER -