Posture-invariant hybrid scaling weight measurement algorithm for live eels

To obtain higher economic benefits, large eel breeding companies classify live eels by weight. Due to their strong mobility and smooth body surface, living are not suitable for traditional mechanical weight measurement. In this study, a sorting machine based on vision was developed, novel method developed obtaining measurements through images. First, backlit workbench designed capture static images of eels, then the projection area skeleton length were obtained image preprocessing. For eel's shape, which is generally cylindrical gradually transitions flat tail, tail posture changes affect shape images; thus, measurement model combining projected proposed. The optimal scale division coefficient found be 0.745 experimentation. Then, select different ranges used error verification correction function error. gradient confirm corrected model. Finally, system calculation results compared with actual results. root mean square (RMSE) 12.94 g, absolute percentage (MAPE) 2.12%. show that proposed provided convenient, fast, low-cost non-contact reduced damage rate can meet technical requirements production. Keywords: factor, measurement, non-contact, DOI: 10.25165/j.ijabe.20231602.7132 Citation: Liu Q, Han Y X, Yan G Mo J S, Yang Z S. Posture-invariant hybrid scaling algorithm eels. Int Agric & Biol Eng, 2023; 16(2): 207-215.