Seat-belt Detection Using Image Processing in MATLAB
Keywords:Seat belt detection, Edge detection, Hough transform, Image closing, Matlab, line detection, image processing
In this study, a method is suggested to detect seat belts in a monitoring image of a moving vehicle. The method relies on several image processing techniques such as edge detection, closing operation, and Hough transform, to precisely identify the seat belts. The proposed method for detecting seat belts is more dependable and sturdier compared to the methods that previously depended on driver's visual monitoring or basic belt detection systems. The initial step of the algorithm involves detecting the lines present in the input image, achieved through the utilization of the Hough transform. Then, the detected lines are filtered based on their orientation and location, allowing only the lines that are likely to represent seat belts to be selected. The closing operation is then used to connect the selected lines and form a complete seat belt. Finally, the algorithm verifies the seat belt by checking the continuity and width of the detected line. The proposed method was evaluated on a dataset of monitoring images containing various driving scenarios, and the results show high accuracy in seat belt detection. The algorithm was able to detect seat belts with a high true positive rate demonstrating its effectiveness in real-world scenarios. Overall, the proposed method provides a reliable and accurate approach to seat belt detection, which can be integrated into existing monitoring systems in vehicles to improve safety. This approach underscores the potential of technology to improve road safety and mitigate the likelihood of accidents, as demonstrated through the application of image processing techniques.
NHTSA. (2022). Seat Belts. [Online]. Available from: https://www.nhtsa.gov/risky-driving/seat-belts
Kashevnik A, Ali A, Lashkov I, Shilov N. Seat belt fastness detection based on image analysis from vehicle in-abin camera. In 2020 IEEE 26th Conference of Open Innovations Association (FRUCT). 2020 Apr 20; 143–150.
Hassanein AS, Mohammad S, Sameer M, Ragab ME. A survey on Hough transform, theory, techniques and applications. arXiv preprint arXiv:1502.02160. 2015 Feb 7.
Zhang DK. An extended opening operation and its application in image processing. In 2008 IEEE International Conference on MultiMedia and Information Technology. 2008 Dec 30; 377–380.
Zhang D. Extended closing operation in morphology and its application in image processing. In 2009 IEEE International Conference on Information Technology and Computer Science. 2009 Jul 25; 1: 83–87.
Salembier P. Structuring element adaptation for morphological filters. J Vis Commun Image Represent. 1992 Jun 1; 3(2): 115–136.
Ziou D, Tabbone S. Edge detection techniques-an overview. Pattern Recognition and Image Analysis C/C of Raspoznavaniye Obrazov I Analiz Izobrazhenii. 1998 Dec;8:537-59.
Mondal R, Dey MS, Chanda B. Image restoration by learning morphological opening-closing network. Mathematical Morphology-Theory and Applications (MMTA). 2020 Jan 1; 4(1): 87–107.
Saravanan C. Color image to grayscale image conversion. In 2010 IEEE 2nd International Conference on Computer Engineering and Applications. 2010 Mar 19; 2: 196–199.
Narendra PM. A separable median filter for image noise smoothing. IEEE Trans Pattern Anal Mach Intell. 1981 Jan; PAMI-3(1): 20–29.