The objective of this work is to perform an analysis of the quality of the transmission of aerial images captured by a UAV and sent in real time to a local server. For this, experiments were carried out with different configurations of transmission quality, varying between high image quality and high fluidity of frames per second (FPS). The drone traveled a distance that varied from 0 to 1000 m (meters), keeping at 50 m high, while the FPS of the transmission were collected and plotted on a graph. The analysis of the results showed that there is a significant difference in human perception and computer perception of the transmitted images. For humans, it is more important to have an image with high quality and contrast, which allows to identify details and colors. For systems involving computer vision, it is more important to have an image with greater fluidity and lower contrast, but that enables to process information faster and avoid delays in communication.
|
