Abstract： Objective To design an automatic measurement method for fetal biparietal diameter (BPD),head circumference(HC)and femoral shaft length(FL)in ultrasound images based on visual features of digital image. Methods The automatic measurement method included three main steps:fetal ultrasound image enhancement,connected domain extraction,and target measurement. The findings of automatic measurement were compared with those of manual annotation,and the maximum error value was recorded. Results A total of 155 images were tested,including 75 of HC and BPD and 80 of femoral shaft. The positioning accuracy of automatic measurement method was more than 90% and 95% for HC and femoral shaft,respectively. The measurement error for fetal HC,BPD and FL was less than 5 mm,2 mm and 0.3 mm,respectively. The t test probability of PHC,PBPD,PFL and α were 0.15,0.13,0.09 and 0.05, respecitvely. The test findings showed that there were no statistically significant differences in BPD,HC and FL between the evaluation of computer aided automatic measurement and manual annotation(all P>0.05). Conclusion The automatic measurement of digital image visual features shows high accuracy and computational efficiency,which can reduce the subjectivity and randomness of manual measurement, improve the accuracy of ultrasound screening,and reduce the workload of medical staff in ultrasound department.