Design and implementation of a real-time LDWS with parameter space filtering for embedded platforms

Selim, Erman; Alcı, Musa; Uğur, Aybars

doi:10.1007/s11554-022-01213-3

Cited by 3 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Finally, some authors have addressed the lane detection topic considering the real-time constraints (Küçükmanisa et al, 2019;Mammeri et al, 2016;Nguyen et al, 2018;Selim et al, 2022), as discussed in the introduction. However, the literature has yet to address heterogeneous computing to improve the computational performance in the matter.…”

Section: Work Related To the Implemented Algorithmmentioning

confidence: 99%

“…Although some authors (Küçükmanisa et al, 2019;Mammeri et al, 2016;Nguyen et al, 2018;Selim et al, 2022) have proposed and evaluated algorithms considering the real-time constraint on embedded systems, none of these, among others, have directly assessed the benefits of a CUDA implementation in an embedded platform. In addition, many implementations use libraries such as OpenCV (Küçükmanisa et al, 2019;Mammeri et al, 2016;Selim et al, 2022). Hence, we contribute to the real-time lane detection topic by implementing and evaluating a CUDA-based solution in a typical low-cost embedded platform.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Real-Time Ego-Lane Detection in a Low-Cost Embedded Platform using CUDA-Based Implementation

Da Silva,

Batista,

Gazzoni Filho

et al. 2023

Semin., Exact Tech. Scienc.

View full text Add to dashboard Cite

This work assesses the effectiveness of heterogeneous computing based on a CUDA implementation for real-time ego-lane detection using a typical low-cost embedded computer. We propose and evaluate a CUDA-optimized algorithm using a heterogeneous approach based on the extraction of features from an aerial perspective image. The method incorporates well-known algorithms optimized to achieve a very efficient solution with high detection rates and combines techniques to enhance markings and remove noise. The CUDA-based solution is compared to an OpenCV library and to a serial CPU implementation. Practical experiments using TuSimple's image datasets were conducted in an NVIDIA's Jetson Nano embedded computer. The algorithm detects up to 97.9% of the ego lanes with an accuracy of 99.0% in the best-evaluated scenario. Furthermore, the CUDA-optimized method performs at rates greater than 300 fps in the Jetson Nano embedded system, speeding up 25 and 140 times the OpenCV and CPU implementations at the same platform, respectively. These results show that more complex algorithms and solutions can be employed for better detection rates while maintaining real-time requirements in a typical low-power embedded computer using a CUDA implementation.

show abstract

Section: Work Related To the Implemented Algorithmmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Real-Time Ego-Lane Detection in a Low-Cost Embedded Platform using CUDA-Based Implementation

Da Silva,

Batista,

Gazzoni Filho

et al. 2023

Semin., Exact Tech. Scienc.

View full text Add to dashboard Cite

show abstract

Construction of Embedded Platform Real-Time Target Tracking System Based on Changing Ant Colony Algorithm

Liu

2023

2023 International Conference on Data Science and Network Security (ICDSNS)

View full text Add to dashboard Cite

Hardware Optimization and System Design of Elliptic Curve Encryption Algorithm Based on FPGA

Gao

2022

Journal of Sensors

View full text Add to dashboard Cite

Since entering the era of big data, the degree of information sharing is getting higher and higher; the information exchange is becoming more and more convenient, but at the same time, personal information is also easy to be exposed to the network environment, if it is used by criminals to lead to information leakage, and then bring certain risks. Therefore, it is in the information age and do a good job of network information security and confidentiality. At present, the security and secrecy of network information are mainly realized by cryptography. Public key cryptography can encrypt information and ensure the security of information transmission, so it is widely used in the contemporary society. At present, elliptic curve encryption is highly respected in the research field of public key cryptosystem. Elliptic curve encryption is divided into two main points, multiplication and inversion, respectively. Through the comparison of these two algorithms, it can be found that there are several choices if the main research objective is to save time, and the Euclidean extension method is mainly discussed in this paper. In other words, more efficient algorithms are used in the hardware implementation process, and a variety of algorithms can be used instead of a single curve algorithm. In this process, we can find the special features of upper level operation and bottom level finite operation. The upper level operation is KP operation, while the bottom level operation is fast calculation of four kinds of K in finite field operation, and finally realize FPGA algorithm. With the help of Quartus ii developed by predecessors, the upper and lower operations of elliptic curve are carried out using VHDL language. Combined ANXIX9.62 in the elliptic curve of each module to test, so as to ensure the accuracy of the data, reduces the error. According to the test results, the designed chip can efficiently complete the elliptic curve encryption system in the whole process. And the average KP operation time can reach 15.15 ms at 20 MHz frequency. At the same time, the chip can complete the operation on ECC public key with any variable curve in F domain less than 256. Therefore, this chip is a high-speed elliptic curve cryptographic chip with optional system parameters. Based on this, this article on the elliptic curve encryption algorithm based on FPGA hardware implementation of system design, from the view of mathematical study analysis, was carried out on the elliptic curve cryptosystem, according to the above two big difficulty, namely, the polynomial of GF(2), the finite field multiplication, and inversion; there will be a detailed studies of discussion, through software comparison to find the differences between different software, especially the software implementation performance level. In addition, it will also focus on the design of elliptic curve algorithm PGA, so as to explore the solution of the algorithm hardware.

show abstract

Design and implementation of a real-time LDWS with parameter space filtering for embedded platforms

Cited by 3 publications

References 21 publications

Real-Time Ego-Lane Detection in a Low-Cost Embedded Platform using CUDA-Based Implementation

Real-Time Ego-Lane Detection in a Low-Cost Embedded Platform using CUDA-Based Implementation

Construction of Embedded Platform Real-Time Target Tracking System Based on Changing Ant Colony Algorithm

Hardware Optimization and System Design of Elliptic Curve Encryption Algorithm Based on FPGA

Contact Info

Product

Resources

About