Abstract

---

The study identified and acquired relevant data for pipeline monitoring model, formulated a mathematical model, simulat ed the model and evaluated it. This is with a view to developing an intelligent information gathering for pipeline monitor ing and security. Data was acquired by taking different aerial images of vandalisation tools and probable vandals using Unmanned Aerial Vehicle (UAV). Surfer 10 application was used to digitize the Nigerian pipeline grid and distribution network to obtain the pipeline coordinates. The itinerary of the UAV for monitoring pipeline was formulated while Sobel edge detection algorithm was engaged with template matching algorithm for vandal detection. Stored image templates of digging equipment around the vicinity of the pipeline were used in a template matching model to detect the presence or absence of digging activities. The different template matching algorithms method such as sum of absolute difference (SAD), sum of squared differences (SDD) and maximum absolute differences (MAD) were used on random images, where best fit results which produced optimal correlation and peak signal to noise ratio (PSNR) values were adopted for object detection and classification. The designed model was simulated using Simulink in MATLAB and evaluation was done by comparing the efficiency of the model. The simulation results showed SSD technique as having the best accuracy with an average value of 5.7065 x E+04 while SAD and MAD techniques have average values of 4.1835 x E+04 and 1.635 x E+04, respectively

Keywords: Unmanned Aerial Vehicle, Pipeline, SAD, SDD, MAD, PSNR, Sobel