目的 为了更直观的评估大豆蛋白液喷雾雾化程度，辅助筛选出有效扇形面积更大、缺陷更少的喷雾，有助于研究喷雾场的雾化均匀性，以期得到均匀一致的大豆蛋白膜。方法 选取合适的阈值后，对粒子图像速度场仪(Particle Image Velocimetry，PIV)拍摄的图像进行二值化处理，划定喷涂有效区域，利用二值化图像计算喷涂雾化场区域在喷涂有效区域内的点位数占比，将数值定义为相对雾化程度，改变稠度系数、喷涂参数，研究相对雾化程度的变化趋势。结果 低稠度系数的大豆蛋白液喷雾相对雾化程度受喷涂参数的影响较小，并保持在90%以上;中高稠度系数的大豆蛋白液喷雾相对雾化程度随着喷涂流量的增加而减小，随着液压、气压的增大有增大的趋势，随着气液比的增加先增大后减小。结论 基于二值化的PIV图像处理可以有效地辅助评估大豆蛋白液喷雾场的均匀性。

In order to more intuitively evaluate the spray atomization degree of soybean protein liquid, the spray situation with larger effective sector area and fewer defects was assisted to screen out, which was conducive to the study of the atomization uniformity of the spray field, in order to obtain uniform soybean protein film. After selecting appropriate thresholds, the images taken by Particle Image Velocimetry (PIV) were binarized to delimit the spraying effective area. The proportion of points in the spraying spraying effective area was calculated by using the binarization Image. The value was defined as the relative atomization degree. The variation trend of relative atomization degree was studied by changing the consistency coefficient and spraying parameters. The relative atomization degree of soybean protein liquid with low consistency was less affected by spraying parameters and remained above 90% . The relative atomization degree of soybean protein liquid spray with medium and high consistency coefficient decreased with the increase of spraying flow, and had a tendency to increase with the increase of, hydraulic pressure and air pressure, and first increased and then decreased with the increase of gas-liquid ratio. PIV image processing based on binarization can effectively assist in evaluating the uniformity of soybean protein liquid spray field.