Edward Clay

Edward Clay

Auto-generated profile, based on publicly available information

Edward Clay

Department of Civil Engineering, California State Polytechnic University, Pomona, CA

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
26 July 2022 in Transportation Research Record: Journal of the Transportation Research Board
Verified
Journal Article
Closed access
LTLong Ngo Hoang Truong
Long Ngo Hoang Truong
ECEdward Clay
Edward Clay
OMOmar Mora
Omar Mora
WCWen Cheng
Wen Cheng
MSMankirat Singh
Mankirat Singh
XJXudong Jia
Xudong Jia
Parking space management systems help organize and optimize available parking spaces for consumers, making the process of finding and using parking spaces more efficient. Current parking space management systems include manual recognition, the employment of magnetic and ultrasonic sensors, and, recently, computer vision (CV). One relatively new region-based convolutional neural network (R-CNN) model, Mask R-CNN, has shown promise in its ability to detect objects and has demonstrated superior performance over many other popular CV methods. Building on Mask R-CNN, an updated version, Rotated Mask R-CNN, which can generate bounding boxes the axes of which are rotated with respect to the image’s axis, was proposed to address the limitation of Mask R-CNN. Albeit with the documented theoretical benefits, the application of the rotated version is rare because of its recent invention. To this end, the study aims to detect vehicle instances in one parking lot using various Rotated Mask R-CNN models based on unmanned aircraft system collected images. Both average precision and average recall were utilized to assess the performance of the alternative models with different backbone and head networks. The results reveal the high accuracy level associated with Rotated Mask R-CNN in real-time detection of vehicles. In addition, the results indicate that the inference speed and total loss are highly correlated with head networks and training schedules.

14 June 2021 in International Journal of Injury Control and Safety Promotion
Verified
Journal Article
Closed access
WCWen Cheng
Wen Cheng
MSMankirat Singh
Mankirat Singh
ECEdward Clay
Edward Clay
JKJerry Kwong
Jerry Kwong
MCMenglu Cao
Menglu Cao
ATAaron Truong
Aaron Truong
Temporal trait of crashes has huge impact on road crash occurrence and a large proportion of research have considered different time periods to determine the causes and features of crash occurrence or frequency. Compared with other safety studies based on a single time interval, considerably less research has relied on the use of multiple time units, especially for the time intervals of less than one year. The research aims to fill the gap by investigating the temporal distribution of crash counts using multiple time spans including hour, weekday and month. To illustrate the most accurate results possible, both the Chi-square test and Cochran–Mantel–Haenzel tests were employed to explore the independence of various time units based on two-way and three-way contingency tables. Eight contingency table models were developed which can be classified into four groups including Complete Independence, Joint Independence, Conditional Independence and Homogeneous Association. Finally, a set of evaluation criteria were utilized for evaluation of the model performance. The results revealed the significant association existence in all time variables (hour, weekday, month) and the model with both main and all interactive effects of time variables provides best prediction performance. Also, the findings showed that Hour 18, weekdays 1, 6, 7 (Friday and Weekends), and month 8 (August) have the largest number of crash occurrences. It is suggested that both main and interactive effects of time variables should be included for model development, which otherwise might yield misleading information. It is anticipated that research results will benefit the safety professionals with better understanding of the temporal patterns of crashes with different time periods and allow the safety administrators to allocate the safety resources.

1 February 2022
Verified
Report
Closed access
PDF file
WCWen Cheng
Wen Cheng
YZYongping Zhang
Yongping Zhang
ECEdward Clay
Edward Clay
Individuals who walk and cycle experience a variety of health and economic benefits while simultaneously benefiting their local environments and communities. It is essential to correctly obtain pedestrian and bicyclist counts for better design and planning of active transportation-related facilities. In recent years, crowdsourcing has seen a rise in popularity due to the multiple advantages relative to traditional methods. Nevertheless, crowdsourced data have been applied in fewer studies, and their reliability and performance relative to other conventional methods are rarely documented. To this end, this research examines the consistency between crowdsourced and traditionally collected count data. Additionally, the research aims to develop the adjustment factor between the crowdsourced and permanent counter data and to estimate the annual average daily traffic (AADT) data based on hourly volume and other predictor variables such as time, day, weather, land use, and facility type. With some caveats, the results demonstrate that the Street Light crowdsourcing count data for pedestrians and bicyclists appear to be a promising alternative to the permanent counters.