نوع مقاله : مقاله پژوهشی
عنوان مقاله English
نویسندگان English
Optimal management of water resources in irrigation networks requires accurate monitoring of hydraulic parameters such as water level and gate opening. Traditional methods are costly, prone to human error, and limited in continuous monitoring. In this research, a framework based on classical image processing and the K-means segmentation algorithm is presented that estimates the flow rate through the gate by identifying the water region beneath defective and worn gates and using the water surface level. The proposed method includes preprocessing, conversion to CIELAB space, dimensionality reduction to 800×600, segmentation with K-means (optimal number of clusters equal to 2, i.e., k=2), morphological refinement, and calculation of geometric parameters. The scale coefficient was determined using OCR and the gate width. Evaluation of the algorithm's application on images showed that the adequacy index was more than 0.992006 and the efficiency index was more than 0.921561, and the root mean square error was 0.010741 under ideal conditions and 0.146896 under severe failures. It also showed that a 10% error in flow rate causes 6.9% and 4.6% errors in adequacy and efficiency, respectively.
کلیدواژهها English