Research on settlement particle recognition based on fuzzy comprehensive evaluation method

• Main Authors: Ran Zhou, Huazhu Song, Jun Li
• Format: Article
• Language: English
• Published: SpringerOpen 2018-10-01
• Series: EURASIP Journal on Image and Video Processing
• Subjects: Settlement particle recognition; Fuzzy partition entropy; Fuzzy comprehensive evaluation method
• Online Access: http://link.springer.com/article/10.1186/s13640-018-0344-0

Abstract In order to solve the problems of the laser scattering rate of settlement particles during sedimentation, such as polymerization, coverage, and disappearance, the grayscale characteristics, morphological features, and motion characteristics of the settlement particles are analyzed and studied in this paper. On the basis of these, the recursive idea is applied to the multi-threshold segmentation algorithm with fuzzy 3-partition entropy algorithm, and then, the fuzzy comprehensive evaluation method is used to identify the settlement particles. Finally, the proposed method is implemented in MATLAB 9 and compared with the traditional Kalman filtering and Otsu segmentation algorithm. The experimental results show that the proposed algorithm is better than other algorithms on the ROC curve, and the recognition rate of the settling particles is higher.