Classification of Potholes using Convolutional Neural Network Model
IEEE BASE PAPER ABSTRACT:
Rough and damaged roads give poor ride quality which leads to poor transport experience, high travel cost, physical loss to vehicles and passengers, and high vehicle maintenance cost. Therefore, to plan a safe and optimal road trip, prior information of road conditions is most important. Poor road conditions are also responsible for traffic accidents. There exist many road conditions and potholes detected techniques and these techniques can be categorized into two basic categories named as vision-based techniques and vibration-based techniques. In this paper vision-based technique is used for road image classification. For the classification, a transfer learning-based Inception ResnetV2 transfer learning based Convolutional Neural Network model is used. The selected model is trained and tested on 3211 road images. These images were collected from the public data available on the internet and captured using the camera. Data is classified into three categories plane road, large pothole, and small pothole. The accuracy of the classification is calculated in terms of precision, recall, F1-score, support, and accuracy percentage. According to performed analysis, the Inception ResnetV2 transfer learning-based Convolutional Neural Network model attained maximum accuracy of 94.42 percent with a 0.933 precision value. The performance of the model during the classification process is evaluated using the training and testing loss. Further, the accuracy of the proposed model is compared with Convolution Neural Network and Support Vector Machine using the same dataset. This paper also provides a comparative analysis of the proposed model with other published work.
PROJECT OUTPUT VIDEO:
OUR PROPOSED ABSTRACT:
The rapid growth of urban infrastructure demands effective and efficient road maintenance strategies. Potholes, a common road infrastructure issue, contribute to increased maintenance costs and safety hazards. In addressing this challenge, the project “Classification of Potholes using Convolutional Neural Network Model” proposes a robust solution utilizing the power of deep learning. The project employs the GoogleNet architecture and is implemented in MATLAB to accurately classify road conditions into three categories: Large Pothole, Small Pothole, and Normal Road. The core of the project lies in the application of Convolutional Neural Networks (CNNs), a subset of deep learning, to automatically classify road images into distinct categories. GoogleNet, a state-of-the-art CNN model, is chosen for its exceptional ability to capture intricate features within images, thereby enhancing the accuracy of classification. The implementation is carried out using the MATLAB programming environment, leveraging its comprehensive image processing capabilities and neural network toolbox. The proposed model is trained on a diverse dataset of road images encompassing various pothole sizes and road conditions. The training process involves iteratively adjusting the model’s parameters to minimize the classification error. By utilizing an extensive dataset and leveraging the capacity of GoogleNet, the model achieves an impressive classification accuracy of 98%. The outcomes of this project are highly promising. The accurate classification of potholes enables municipal authorities and road maintenance teams to swiftly identify road sections in need of repair, prioritize maintenance efforts, and allocate resources effectively. Additionally, the automated classification approach reduces the dependence on manual inspections, resulting in substantial time and cost savings. The proposed “Classification of Potholes using Convolutional Neural Network Model” demonstrates a novel and effective approach to addressing road maintenance challenges. Through the utilization of the GoogleNet architecture within the MATLAB environment, the project achieves a remarkable classification accuracy of 98%, paving the way for enhanced road infrastructure management and safer driving conditions.
- The primary objective is to develop a robust and efficient deep learning framework capable of accurately detecting and categorizing different types of potholes in images.
- This framework has the potential to revolutionize road maintenance strategies, enhance safety, and optimize resource allocation for infrastructure repair.
- System : Pentium i3 Processor.
- Hard Disk : 500 GB.
- Monitor : 15’’ LED.
- Input Devices : Keyboard, Mouse.
- Ram : 8 GB.
- Operating system : Windows 10 Pro.
- Coding Language : MATLAB
- Tool : MATLABR2021A
Saravjeet Singh; Rishu Chhabra; Aditi Moudgil, “Classification of Potholes using Convolutional Neural Network Model: A Transfer Learning Approach using Inception ResnetV2”, IEEE Conference, 2023.