UROP Proceedings 2020-21

School of Engineering Department of Computer Science and Engineering 98 AI meets Big Data: User Analytics and Personalized Recommendation Based on Location Data Supervisor: CHAN Gary Shueng Han / CSE Student: PENG Zhuoxuan / COSC Course: UROP2100, Fall License plate recognition (LPR) is an important computer vision task in transportation and vehicle management. In our project, LPR is performed in real time on Raspberry Pi, a power efficient device. This requires a series of techniques to reduce the computational resources and boost the recognition speed. This semester I first finished the web streaming system, then contributed to the development of processing of pictures and recognition. The major achievement is the implementation of a calibration algorithm by picking out only areas possibly containing license plates. Next I will focus on the connection of different systems in this project. AI meets Big Data: User Analytics and Personalized Recommendation Based on Location Data Supervisor: CHAN Gary Shueng Han / CSE Student: XIANG Letian / COMP Course: UROP2100, Fall With the development of digital economy, huge amounts of geographical data that depicts paths from an origin to a destination are created daily. They are generated to businesses like ride-hailing services (for example, Uber and Didi), food delivery services (for example, FoodPanda, DoorDash and Meituan), and express delivery services (for example, FedEx and USPS). Predicting the data for different origin-destination pairs is essential for facilitating these services’ qualities. To find an effective solution to this problem, we need to make use of spatial-temporal series to find the correlations with deep learning techniques. We will also show the result of studies of some model from previous works on traffic prediction from spatialtemporal similarity. AI meets Big Data: User Analytics and Personalized Recommendation Based on Location Data Supervisor: CHAN Gary Shueng Han / CSE Student: ZHANG Daofu / COMP Course: UROP2100, Fall In this project, we are aiming to develop a system which can be planted in RPi to automatically recognize license plate information from given images of automobiles. This can help people to manage the car information in the parking lot. There are mainly two steps, first we need to locate the place of the license plate from the given images which are usually taken from cameras in the parking lots. After we obtained the image of the license plate, we used another model to recognize the symbols in it and produce the final result.

RkJQdWJsaXNoZXIy NDk5Njg=