In this paper, we present a novel approach to ray mapping and detection through the development of machine vision algorithms. The primary objective is to enhance the efficiency and accuracy of identifying and locating out-of-control radioactive sources in complex and dynamic environments. Initially, an algorithm was devised based on the movement patterns of Timio's wheeled robots. This algorithm takes into account various factors, such as their random movement within the environment, obstacle and wall detection capabilities, directional adjustments when approaching each other, and other relevant scenarios. To facilitate experimentation, we defined 10 representative characters of these robots on the following page, each occupying a 10 x 10 square meter area. A video capturing their movement was recorded over a duration of 120 seconds at a frame rate of 25 frames per second. The coordinates of their movement paths were then recorded within this time frame. Subsequently, a machine vision algorithm was developed based on the KLT tracking method equations. This algorithm effectively tracked the movement paths of the characters, and the resulting coordinates were compared with the actual movement coordinates for validation. In the next phase, a radiation scenario was introduced by placing a radioactive source on one of the characters. To simulate this, we employed 3000 Monte Carlo codes specifically designed to account for the presence of the source. The output of these codes, recorded as counts in the detector, was extracted and stored for analysis. Finally, to detect the moving radioactive source within an environment containing a high number of characters, algorithms based on data correlation between the movement paths and the recorded counts in the detector were utilized.
Beigzadeh A, Ardiny H. Radiation mapping and detection of out-of-control radioactive sources by developing algorithms based on machine vision. JGST 2025; 15 (2) : 5 URL: http://jgst.issgeac.ir/article-1-1176-en.html