2025_proceedings: Enhancing the Bearing Time Record Image Based on Human Vision

• Session: Poster session
  Organised by: N/A
• Lecture: Enhancing the Bearing Time Record Image Based on Human Vision
  Paper ID: 2142
  Author(s): Yi LI, Hua FANG, Enming ZHENG, Chunnan SONG
  Presenter: Yi LI
  Abstract: To improve the performance of the sonar’s detection capabilities, an algorithm which is based on the human vision has been introduced to enhance the bearing time record (BTR) image which is the output of the towed line array. It is well known that the most rudimentary aspect of the human vision is contrast, hence the concept of contrast was presented first in this paper. Previously, it was first proposed that the crucial tactics for detecting weak targets in an image is to reduce the global contrast of the image and increase its local contrast. Then, a simplified model of beam pattern was introduced. It was then shown that the gray-scale transformation could be used to reduce the global contrast of the image, and the image sharpening could be used to increase its local contrast. Lastly, contrast stretching was used to adjust the image to the appropriate contrast range. In this paper, the performances of this algorithm such as the variation of the brightness of the targets and that of the bearing of the targets were analyzed and discussed. Analysis of the sea trial data has shown that the invisible targets in the original BTR image appeared after the processing of the algorithm which can be implemented in real-time and in practice.
• Corresponding author: Dr Yi LI
  Affiliation: Institute of Acoustics, Chinese Academy of Sciences
  Country: China