Real-time concrete crack detection and instance segmentation using deep transfer learning
- Authors: Piyathilaka, Lasitha , Preethichandra, Daluwathu , Izhar, Umer , Appuhamillage, Gayan
- Date: 2020
- Type: Text , Journal article
- Relation: Engineering Proceedings Vol. 2, no. 1 (2020), p.
- Full Text:
- Reviewed:
- Description: Cracks on concrete infrastructure are one of the early indications of structural degradation which needs to be identified early as possible to carry out early preventive measures to avoid further damage. In this paper, we propose to use YOLACT: a real-time instance segmentation algorithm for automatic concrete crack detection. This deep learning algorithm is used with transfer learning to train the YOLACT network to identify and localize cracks with their corresponding masks which can be used to identify each crack instance. The transfer learning techniques allowed us to train the network on a relatively small dataset of 500 crack images. To train the YOLACT network, we created a dataset with ground-truth masks from images collected from publicly available datasets. We evaluated the trained YOLACT model for concrete crack detection with ResNet-50 and ResNet-101 backbone architectures for both precision and speed of detection. The trained model achieved high mAP results with real-time frame rates when tested on concrete crack images on a single GPU. The YOLACT algorithm was able to correctly segment multiple cracks with individual instance level masks with high localization accuracy.
Delivery of online electronics and mechatronics labs during lockdowns
- Authors: Jayawardena, Amal , Kahandawa, Gayan , Piyathilaka, Lasitha
- Date: 2021
- Type: Text , Conference paper
- Relation: 8th IEEE International Conference on e-Learning in Industrial Electronics, ICELIE 2021, Virtual, Toronto, 13-16 October 2021, Proceedings - 2021 8th IEEE International Conference on e-Learning in Industrial Electronics, ICELIE 2021
- Full Text: false
- Reviewed:
- Description: This paper provides a detailed explanation of several approaches that can be used to conduct online labs for elec-tronics/mechatronics engineering courses and explains the results obtained from a survey conducted. The detailed explanations provide information on how to implement the method, benefits of the stated process, possible challenges, and how to overcome those challenges. Furthermore, this paper presents the analyzed results from a survey conducted to capture the student experience in online labs. © 2021 IEEE.
Enhanced learner interactions and academic integrity with bespoke interactive online tutorials in a hybrid learning model
- Authors: Sul, Jay , Piyathilaka, Lasitha , Tahmoorian, Farzaneh , Kahandawa, Gayan , Gudimetla, Prasad
- Date: 2024
- Type: Text , Conference paper
- Relation: 8th IEEE World Engineering Education Conference, EDUNINE 2024, Hybid, Guatemala City, 10-13 March 2024, EDUNINE 2024 - 8th IEEE World Engineering Education Conference: Empowering Engineering Education: Breaking Barriers through Research and Innovation, Proceedings
- Full Text: false
- Reviewed:
- Description: Conventional teaching methods have morphed into blended and hybrid learning models with the evolution of online platforms. These have gained importance and popularity during and post-COVID-19. While the previous studies reported great success in this transition, they are mainly oriented around flexibility and learner interactions with online materials, instead of interactions with instructors. This paper discusses the development and implementation of interactive online tutorials to enhance hybrid learning in a first-year engineering subject over six years. These tutorials were designed not only to replicate the benefits of face-to-face tutorial classes, but also to enhance student interactions and engagement with teaching staff, and to detect potential academic misconduct. This study found that the interactive online tutorials (i) provided a motivational and positive learning experience to all students, (ii) narrowed the gap in satisfaction levels between on-campus and online students, and (iii) benefited the higher academic achievers within the online student cohorts. © 2024 IEEE.
Navigating the new normal: student perspectives on transitioning from online to face-to-face learning after COVID-19 lockdowns
- Authors: Jayawardena, Amal , Kahandawa, Gayan , Hewawasam, Hasitha , Piyathilaka, Lasitha , Sul, Jay
- Date: 2024
- Type: Text , Conference paper
- Relation: 8th IEEE World Engineering Education Conference, EDUNINE 2024, Hybid, Guatemala City, 10-13 March 2024, EDUNINE 2024 - 8th IEEE World Engineering Education Conference: Empowering Engineering Education: Breaking Barriers through Research and Innovation, Proceedings
- Full Text: false
- Reviewed:
- Description: This paper explores the transition from online learning to face-to-face learning in the aftermath of the COVID-19 pandemic. Based on a survey conducted among students in engineering classes, the study investigates the challenges and preferences experienced during this critical period. The survey responses provide valuable insights into lecture delivery methods, time management, social skills, workload comparisons, and the support required for a successful transition. The findings highlight the preference for a hybrid approach that combines the benefits of both online and face- to- face learning. Flexibility in scheduling, access to digital resources, and personalized learning experiences emerged as key factors influencing student satisfaction. Additionally, the survey identifies the need for effective time management strategies, social skills development, and mental health support during the transition. By prioritizing student needs and preferences, educational institutions can create a supportive and engaging learning environment that promotes academic success and well-being in the post-pandemic education landscape. © 2024 IEEE.