Loading...
Please wait, while we are loading the content...
Similar Documents
Live Digit Recognition using Nvidia ’ s Jetson Tx 1
| Content Provider | Semantic Scholar |
|---|---|
| Author | Mercurio, Jayson Paul Yamada, Kevin Choi, Alexander K. Y. Iqbal, Ayesha José, Ignacio Guzmán Enriquez, Amelito G. Zhang, Xiaorong Pong, Wenshen Jiang, Zhaoshuo Chen, Cheng Mahmoodi, H. R. Jiang, Hao |
| Copyright Year | 2018 |
| Abstract | Community colleges provide a beneficial foundation for undergraduate education in STEM majors. To inspire community college students to pursue a major in STEM, it is crucial to adapt strategies that help facilitate this interest. With support from the Department of Education Minority Science and Engineering Improvement program (MSEIP) and the Hispanic-Serving Institution Science, Technology, Engineering and Mathematics (HSI STEM), an internship program with multiple colleges was developed between community colleges and a public fouryear university to engage community college students in cutting-edge engineering research. In the summer of 2017, four community college students participated in a ten-week electrical and computer engineering research internship project at a four-year university research lab. The summer internship project aimed to develop a real-time handwritten digit recognition system leveraging Neural Networks and Nvidia’s Jetson Tx1 platform. Utilizing a modified Nvidia workflow, a robust digit recognition algorithm was designed using two industry standard programs for deep learning -TensorFlow and DIGITS. Nvidia’s live image recognition demonstration created the framework to interface a camera module that sends images to the input of the digit classifying network in real-time. The student interns designed experiments to test the robustness of the algorithm in their daily environment, from low light situations to cluttered backgrounds with the handwritten digit blending in. The internship project created a stimulating environment for student interns to gain research experiences and learn a wealth of knowledge in deep learning, real time pattern recognition systems and leading-edge hardware platforms. The experiences contained within the ten-week internship allowed the interns to drastically improve technical writing and presentations, experimental design, data analysis and management, teamwork, and perseverance. The ten-week research internship was an effective method for engaging aspiring community college students by teaching the tools and methodology for success within an engineering profession, and helping to increase the interns’ confidence levels. |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://peer.asee.org/inspiring-community-college-students-in-electrical-and-computer-engineering-research-through-live-digit-recognition-using-nvidia-s-jetson-tx1.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
