부산대학교 도서관

The measurement of morphometric parameters of fish erythrocytes (MFEs) has been recognized as a valuable tool for ecological biomonitoring. Conventional methods for measuring MFE greatly depend on manual measurements, leading to a labor-intensive and time-consuming process that relies on human expertise. To overcome these issues, in this article, a solution based on image processing and ellipse fitting to provide an automated framework for the measurement of MFE is presented. The performance of the proposed method is assessed quantitatively by comparing its results with manual measurements performed by an expert biologist. In addition, a comparison of the proposed method with three state-of-the-art methods is presented by providing the statistical analysis of their mean absolute errors. The analysis of results demonstrates the proposed method significantly outperforms existing methods for both cell and nucleus measurement. It notably achieves a lower absolute mean error value of 78 pixels for cell area and 36 pixels for nucleus area, confirming the better performance of the proposed method compared with the others. The experimental results demonstrate that the proposed method can correctly fit the ellipse on both fish blood cells and their nuclei.