Inexpensive Multimodal Sensor Fusion System for Autonomous Data Acquisition of Road Surface Conditions

Chen, Yulu Luke; Jahanshahi, Mohammad R.; Manjunatha, P.; Gan, WeiPhang; Abdelbarr, Mohamed; Masri, Sami F.; Becerik-Gerber, Burçin; Caffrey, John

doi:10.1109/jsen.2016.2602871

Cited by 49 publications

(19 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on the initial results, it was observed that using C or C++ language has advantage over MATLAB or Python for efficient and fast data acquisition since, for instance, some image frames were missed when Python was used for data acquisition. This phenomenon was also observed by Chen et al [24] when they used MATLAB for collecting RGB and depth data. They concluded that using C or C++ resolves this issue.…”

Section: B Discussion and Advisor Observationssupporting

confidence: 70%

Integrating data information literacy into a service-learning engineering design course

Phillips

Jahanshahi

2017

2017 IEEE Frontiers in Education Conference (FIE)

View full text Add to dashboard Cite

Over the past several years increasing emphasis has been placed on the development of data information literacy (DIL) skills for engineering researchers to comply with funding requirements and to facilitate data sharing, reuse, and preservation. These efforts have been largely aimed at faculty members and graduate students, and to some extent undergraduate students in preparation for their roles as future researchers. However, little attention has been directed at developing DIL skills in undergraduate engineering students engaged in non-research activities, such as service-learning design projects that integrate instruction and community service. This work-in-progress paper reports on using service-learning pedagogy to integrate DIL into engineering design education. Preliminary analysis of student journal entries and instructor observations are discussed.

show abstract

Section: B Discussion and Advisor Observationssupporting

confidence: 70%

Integrating data information literacy into a service-learning engineering design course

Phillips

Jahanshahi

2017

2017 IEEE Frontiers in Education Conference (FIE)

View full text Add to dashboard Cite

show abstract

“…The camera position is not important in this study because the aim was developing a road profile rather than presenting the pavement images on a map. Related literature shows that, by fusing the GPS and IMU data, location data with a higher frequency can be generated [42]. In addition, since a GPS dataset creates timestamps on captured images in a sequence, the images can be labeled with time and location [49].…”

Section: Methodsmentioning

confidence: 99%

“…Chen et al (2016) used multiple Kinect sensors to evaluate the road surface condition in a dynamic mode. Having solved the problems related to data collection at high speed, such as the motion blur problem and the rolling shutter distortion, they detected the surface defects by running the equipped vehicle at the traffic speed limit [42]. Zhang et al (2018) detected and classified the intensity of different kinds of cracks by measuring their widths, lengths, and areas.…”

Section: Literature Reviewmentioning

confidence: 99%

Estimating Pavement Roughness by Fusing Color and Depth Data Obtained from an Inexpensive RGB-D Sensor

Mahmoudzadeh

Golroo

Jahanshahi

et al. 2019

Sensors

View full text Add to dashboard Cite

Measuring pavement roughness and detecting pavement surface defects are two of the most important tasks in pavement management. While existing pavement roughness measurement approaches are expensive, the primary aim of this paper is to use a cost-effective and sufficiently accurate RGB-D sensor to estimate the pavement roughness in the outdoor environment. An algorithm is proposed to process the RGB-D data and autonomously quantify the road roughness. To this end, the RGB-D sensor is calibrated and primary data for estimating the pavement roughness are collected. The collected depth frames and RGB images are registered to create the 3D road surfaces. We found that there is a significant correlation between the estimated International Roughness Index (IRI) using the RGB-D sensor and the manual measured IRI using rod and level. By considering the Power Spectral Density (PSD) analysis and the repeatability of measurement, the results show that the proposed solution can accurately estimate the different pavement roughness.

show abstract

“…Focusing solely on the aspect ratio (width to height) a square is distinguished from all other primitive forms when the ratio is equal to one (= 1.0): (11) Although relatively simple, identifying shapes in this manner raises a complex problem. If the square metrics and the circle metrics return a value close to 1 the shapes could be classified incorrectly.…”

Section: Output Classmentioning

confidence: 99%

“…RBC denotes an ability to break complex images into a series of primitive forms such as square, triangle, circle or rectangle. Cross-correlating this information with the arrangement of the geometric primitive improves the accuracy [11], [12]. This classification method -best viewed as reducing equivocation -relies on increasing the verity of data the machine is exposed to.…”

Section: Introductionmentioning

confidence: 99%

A Cognitive Framework for Object Recognition with Application to Autonomous Vehicles

Roche

Silva

Kondoz

2018

Advances in Intelligent Systems and Computing

View full text Add to dashboard Cite

Autonomous vehicles or self-driving cars are capable of sensing the surrounding environment so they can navigate roads without human input. Decisions are constantly made on sensing, mapping and driving policy using machine learning techniques. Deep Learningmassive neural networks that utilize the power of parallel processinghas become a popular choice for addressing the complexities of real time decision making. This method of machine learning has been shown to outperform alternative solutions in multiple domains, and has an architecture that can be adapted to new problems with relative ease. To harness the power of Deep Learning, it is necessary to have large amounts of training data that are representative of all possible situations the system will face. To successfully implement situational awareness in driverless vehicles, it is not possible to exhaust all possible training examples. An alternative method is to apply cognitive approaches to perception, for situations the autonomous vehicles will face. Cognitive approaches to perception work by mimicking the process of human intelligencethereby permitting a machine to react to situations it has not previously experienced. This paper proposes a novel cognitive approach for object recognition. The proposed cognitive object recognition algorithm, referred to as Recognition by Components, is inspired by the psychological studies pertaining to early childhood development. The algorithm works by breaking down images into a series of primitive forms such as square, triangle, circle or rectangle and memory based aggregation to identify objects. Experimental results suggest that Recognition by Component algorithm performs significantly better than algorithms that require large amounts of training data.

show abstract

Inexpensive Multimodal Sensor Fusion System for Autonomous Data Acquisition of Road Surface Conditions

Cited by 49 publications

References 26 publications

Integrating data information literacy into a service-learning engineering design course

Integrating data information literacy into a service-learning engineering design course

Estimating Pavement Roughness by Fusing Color and Depth Data Obtained from an Inexpensive RGB-D Sensor

A Cognitive Framework for Object Recognition with Application to Autonomous Vehicles

Contact Info

Product

Resources

About