This paper proposes a method for analyzing the video of an unmanned aerial vehicle broadcast under low signal conditions. The low signal level was achieved by increasing the distance between the aircraft and the video data receiver in the form of first-person view glasses. The proposed method for analyzing video data allows detecting the presence of delays, recording distortions in a video frame, and tracking dynamic changes in the video. The proposed method is based on the structural similarity index metric and its various variations with attention to its components, such as contrast, brightness, and pixel structure of the image. Analysis of each aspect of the structural similarity index metric revealed that more significant changes between frames occurred in the aspect of image structure. A comparative analysis was carried out with the metric for assessing the quality of images and videos, which do not require the presence of a reference original image, unlike the structural similarity index metric. Comparison of methods demonstrated the advantage of the proposed method. The proposed method can be applied to solve problems related to image registration by unmanned aerial vehicles, for example, in agriculture, construction, cartography and geology.

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Video Analysis of an Unmanned Aerial Vehicle in Low Signal Conditions

  • Pavel Lyakhov,
  • Anzor Orazaev,
  • Nikolay Nagornov

摘要

This paper proposes a method for analyzing the video of an unmanned aerial vehicle broadcast under low signal conditions. The low signal level was achieved by increasing the distance between the aircraft and the video data receiver in the form of first-person view glasses. The proposed method for analyzing video data allows detecting the presence of delays, recording distortions in a video frame, and tracking dynamic changes in the video. The proposed method is based on the structural similarity index metric and its various variations with attention to its components, such as contrast, brightness, and pixel structure of the image. Analysis of each aspect of the structural similarity index metric revealed that more significant changes between frames occurred in the aspect of image structure. A comparative analysis was carried out with the metric for assessing the quality of images and videos, which do not require the presence of a reference original image, unlike the structural similarity index metric. Comparison of methods demonstrated the advantage of the proposed method. The proposed method can be applied to solve problems related to image registration by unmanned aerial vehicles, for example, in agriculture, construction, cartography and geology.