An Automated Machine-Learning Approach for Road Pothole Detection Using Smartphone Sensor Data

Wu, Chao; Wang, Zhen; Hu, Jun; Lépine, Julien; Na, Xiaoxiang; Ainalis, Daniel; Stettler, Marc

doi:10.3390/s20195564

Cited by 70 publications

(29 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As shown in Table 7 , most studies attempted to utilize RGB images and acceleration data for recognizing road conditions. However, studies from [ 17 , 18 , 19 , 20 ] mainly focused on detecting the damages of a motorcar road and required a smartphone to be installed on the dashboard of a vehicle, thus the method is not suitable for wheelchair users. In contrast, Watanabe et al and Iwasawa et al [ 21 , 22 ] tried to recognize the status of the sidewalk by using acceleration data.…”

Section: Discussionmentioning

confidence: 99%

“…With a recent growth of computer vision and machine learning technologies, there have been various attempts to automatically detect and report defects on roads and sidewalks. Previous approaches primarily captured RGB road images or sensor data (e.g., accelerometer and gyroscope) and exploited deep learning and machine learning algorithms for both detecting road cracks/potholes [ 17 , 18 , 19 , 20 ] and recognizing sidewalk anomalies [ 21 , 22 ]. These methods can automatically detect the defects on the road surface but still have the following limitations: (1) the captured RGB images are not helpful to classify the road condition under low-light conditions (e.g., nighttime) and (2) sensors can produce noisy data or restrict the user’s natural movements, adversely affecting the overall performance.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Classification of the Sidewalk Condition Using Self-Supervised Transfer Learning for Wheelchair Safety Driving

Yoon

Kim

Jeong

2022

Sensors

View full text Add to dashboard Cite

The demand for wheelchairs has increased recently as the population of the elderly and patients with disorders increases. However, society still pays less attention to infrastructure that can threaten the wheelchair user, such as sidewalks with cracks/potholes. Although various studies have been proposed to recognize such challenges, they mainly depend on RGB images or IMU sensors, which are sensitive to outdoor conditions such as low illumination, bad weather, and unavoidable vibrations, resulting in unsatisfactory and unstable performance. In this paper, we introduce a novel system based on various convolutional neural networks (CNNs) to automatically classify the condition of sidewalks using images captured with depth and infrared modalities. Moreover, we compare the performance of training CNNs from scratch and the transfer learning approach, where the weights learned from the natural image domain (e.g., ImageNet) are fine-tuned to the depth and infrared image domain. In particular, we propose applying the ResNet-152 model pre-trained with self-supervised learning during transfer learning to leverage better image representations. Performance evaluation on the classification of the sidewalk condition was conducted with 100% and 10% of training data. The experimental results validate the effectiveness and feasibility of the proposed approach and bring future research directions.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Classification of the Sidewalk Condition Using Self-Supervised Transfer Learning for Wheelchair Safety Driving

Yoon

Kim

Jeong

2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Terrible road conditions such as rough roads, potholes, cracks, manholes, speed bumps, ditches, and surface height imbalance are major sources of vehicle crashes and high death rates. The concrete material quality, large rate of traffic flow, heavy vehicles, and climate changes such as snowfall and heavy rains are affecting road surfaces, and as a result, road anomalies are increasing day by day [ 7 ]. Road surface anomalies are becoming an increasingly important issue for roads around the world, such as potholes and cracks.…”

Section: Introductionmentioning

confidence: 99%

Edge AI-Based Automated Detection and Classification of Road Anomalies in VANET Using Deep Learning

Bibi

Saeed

Zeb

et al. 2021

Computational Intelligence and Neuroscience

121

View full text Add to dashboard Cite

Road surface defects are crucial problems for safe and smooth traffic flow. Due to climate changes, low quality of construction material, large flow of traffic, and heavy vehicles, road surface anomalies are increasing rapidly. Detection and repairing of these defects are necessary for the safety of drivers, passengers, and vehicles from mechanical faults. In this modern era, autonomous vehicles are an active research area that controls itself with the help of in-vehicle sensors without human commands, especially after the emergence of deep learning (DNN) techniques. A combination of sensors and DNN techniques can be useful for unmanned vehicles for the perception of their surroundings for the detection of tracks and obstacles for smooth traveling based on the deployment of artificial intelligence in vehicles. One of the biggest challenges for autonomous vehicles is to avoid the critical road defects that may lead to dangerous situations. To solve the accident issues and share emergency information, the Intelligent Transportation System (ITS) introduced the concept of vehicular network termed as vehicular ad hoc network (VANET) for achieving security and safety in a traffic flow. A novel mechanism is proposed for the automatic detection of road anomalies by autonomous vehicles and providing road information to upcoming vehicles based on Edge AI and VANET. Road images captured via camera and deployment of the trained model for road anomaly detection in a vehicle could help to reduce the accident rate and risk of hazards on poor road conditions. The techniques Residual Convolutional Neural Network (ResNet-18) and Visual Geometry Group (VGG-11) are applied for the automatic detection and classification of the road with anomalies such as a pothole, bump, crack, and plain roads without anomalies using the dataset from different online sources. The results show that the applied models performed well than other techniques used for road anomalies identification.

show abstract

“…As the mobile users (or crowd workers) are equipped with sensing devices, so the researchers thought of utilizing these mobile users for sensing and collecting data for several real world applications and then distributing it to the community or organization. For example, measuring the air pollution level across the cities Pan et al (2017); Samulowska et al (2021); Li et al (2019), giving information about the road traffic Staniek (2021); Aubry et al (2014), noise pollution assessment Pődör and Szabó (2021); Rana et al (2010), information about the potholes Enigo et al (2016); Wu et al (2020), and many more Nagatani et al (2013); Poblet et al (2014). The process of completion of task(s) by the crowd workers or group of common people equipped with sensing devices in the form of an open call give rise to a new paragmatic field of study termed as mobile crowdsensing (a.k.a.…”

Section: Introductionmentioning

confidence: 99%

QUAD: A Quality Aware Multi-Unit Double Auction Framework for IoT-Based Mobile Crowdsensing in Strategic Setting

Singh¹,

Jasti²,

Singh³

et al. 2022

Preprint

View full text Add to dashboard Cite

Crowdsourcing with the intelligent agents carrying smart devices is becoming increasingly popular in recent years. It has opened up meeting an extensive list of real life applications such as measuring air pollution level, road traffic information, and so on. In literature this is known as mobile crowdsourcing or mobile crowdsensing. In this paper, the discussed set-up consists of multiple task requesters (or task providers) and multiple IoT devices (as task executors), where each of the task providers is having multiple homogeneous sensing tasks. Each of the task requesters report bid along with the number of homogeneous sensing tasks to the platform. On the other side, we have multiple IoT devices that reports the ask (charge for imparting its services) and the number of sensing tasks that it can execute. The valuations of task requesters and IoT devices are private information, and both might act strategically. One assumption that is made in this paper is that the bids and asks of the agents (task providers and IoT devices) follow decreasing marginal returns criteria. In this paper, a truthful mechanism is proposed for allocating the IoT devices to the sensing tasks carried by task requesters, that also keeps into account the quality of IoT devices. The mechanism is truthful, budget balance, individual rational, computationally efficient, and prior-free. The simulations are carried out to measure the performance of the proposed mechanism against the benchmark mechanisms. The code and the synthetic data are available at https://github.com/Samhitha-Jasti/QUAD-Implementation.

show abstract

An Automated Machine-Learning Approach for Road Pothole Detection Using Smartphone Sensor Data

Cited by 70 publications

References 40 publications

Classification of the Sidewalk Condition Using Self-Supervised Transfer Learning for Wheelchair Safety Driving

Classification of the Sidewalk Condition Using Self-Supervised Transfer Learning for Wheelchair Safety Driving

Edge AI-Based Automated Detection and Classification of Road Anomalies in VANET Using Deep Learning

QUAD: A Quality Aware Multi-Unit Double Auction Framework for IoT-Based Mobile Crowdsensing in Strategic Setting

Contact Info

Product

Resources

About