Enhancing Water Body Identification in Satellite Images through Comparative Analysis of Canny Edge Detection and Scale-Invariant Feature Transform Algorithms

Abstract

Aim: The objective of this research is to utilize the Innovative Canny Edge Detection algorithm to precisely identify water bodies in satellite images acquired from the Water Access program. Subsequently, the performance of this approach will be evaluated by comparing it with the SIFT algorithm. The primary objective of this analysis is to contribute to the equilibrium of the water ecosystem. Materials and Methods: Within this study, two distinct groups were compared: the Innovative Canny Edge Detection algorithm (CED) (N=10) and the scale invariant feature transform algorithm (SIFT) (N=10). The total sample size was determined with the help of (g) power technology, considering an alpha of 0.05, an enrollment ratio of 1, a 93% confidence interval, and a power of 80%. Results: The Innovative Canny edge detection algorithm exhibited a noteworthy detection rate, achieving an accuracy of 85% compared to the SIFT algorithm's accuracy of 82%. The achieved results were devoid of errors, as indicated by the accuracy of the significance value (0.0001), substantiated by an 80% pre-test power in SPSS Statistical analysis. A significant statistical distinction between the two algorithms was observed. Conclusion: The Innovative Canny Edge Detection algorithm outperforms the SIFT algorithm with statistical significance.