In view of the diversity of behavior and the complexity of influence relationship in measurement,a new strategy of visual measurement for bench drill spindle based on behavior self-adaption was proposed.A visual measurement system of spindle diameter was established by using the combination sampling method of circumferential division position acquisition image and axial segmentation contour to extract the measured information.On this basis,the behavioral adaptive visual measurement process,which integrated functions such as adaptive optimizing exposure interval,batch processing images and error feedback closed-loop measurement,was designed and the corresponding software was developed.Finally,taking a certain type of bench drill spindle as example,the adaptive optimization experiment of exposure interval was carried out,as well as the error analysis and stability evaluation for multi-station,multi-axis and multi-scheme measurement.The results show that the proposed behavior adaptive measurement strategy forms the closed-loop measurement by obtaining the influence of measurement behavior change on measurement error,by which the measurement stability can be effectively improved.It not only meets the needs of high precision measurement of drilling spindle,but also promotes the development of measurement intelligence.