Motorcycle image analysis with nanoelectronics platform

This research investigated of image tracking and edge detection for motorcycle in various lighting and weather conditions with nanoelectronics platform. The platform capability in different resolution and threshold level also evaluated. Comparison between field programmable gate array (FPGA) hardware platform and (MATLAB) software platform has been made. Developed framework showed great accuracy in segmentation of motorcycle plate number in daylight compared to rainy daylight and night condition. System developed has the processing time less than 40 milliseconds in various critical conditions such as daylight, rainy daylight and night conditions. The output image was analyzed by comparing the accuracy of bounding box and edges which is displayed in different conditions, threshold level, resolutions and distances. The result showed different performance for each condition. The output image is clear with pixel 1024 x768 in daylight, rainy and night. Meanwhile image quality getting blur while used low pixel resolutions such as 640 x 480, 720 x 480 and 800 x 600. Total speed for each image processing is 30 frames per second. The ability of this system captured motorcycle image is 5 to 15 meter in daylight, rainy and night. Analysis with Histogram level and contrast stretching method showed performance in hardware is improved rather than software [1-6].

Text 20.pdf Restricted to Repository staff only Download (188kB) Official URL or Download Paper: http://www.samn2018.upm.edu.my/doc/20.pdf

Actions (login required)

View Item