Fig. 1. Traditional Faster RCNN model

Fig. 2. ResNet101 architecture model

Fig. 3. Residual module

Fig. 4. FPN framework

Fig. 5. Network structure of ResNet101 and FPN

Fig. 6. Network framework of channel attention module

Fig. 7. Spatial attention module network framework

Fig. 8. CBAM network framework

Fig. 9. Network structure of improved Faster RCNN

Fig. 10. Improved model water droplet recognition effect

Fig. 11. Water contact angle calculation flow

Fig. 12. Canny edge detection

Fig. 13. Harris corner extraction

Fig. 14. IRLS contour fitting

Fig. 15. The right water contact angle calculated by establishing the coordinate system with the center of the ellipse

Fig. 16. Comparison of location results of droplet region. (a)‒(f) The detection results of the original Faster RCNN; (g)‒(l) the detection result of improved Faster RCNN

Fig. 17. Comparison of the original network mAP curve (thin) and the improved Faster RCNN network mAP curve (thick)

Fig. 18. Image of contact angle changing over time