Fig. 1. (a) Mechanical structure diagram of the proposed erector; (b) Layout of the detection system; (c) Physical drawing and dimension drawing of groove (Unit: mm)

Fig. 2. Segment surface feature extraction framework based on two stage deep neural network

Fig. 3. Schematic diagram of network framework in image restoration stage

Fig. 4. (a) PSNR curve of the first stage; (b) Mask loss function curve of the second stage

Fig. 5. Comparison of different contour feature extraction methods. (a) and (b) Processing results based on the traditional Mask R-CNN algorithm; (c) and (d) Processing results based on proposed algorithm in the paper; (e) and (f) Processing results of original image based on proposed algorithm in the paper

Fig. 6. (a) CE value in X direction; (b) CE value in Y direction

Fig. 7. Experimental diagram of segment automatic assembly positioning

Fig. 8. Change curves of the mark coordinates detected by two cameras with the pose adjustment of the assembly machine during the process of camera grabbing and positioning

Fig. 9. Absolute value of grab positioning error

Fig. 10. Change curves of mark coordinates detected by two cameras with pose adjustment of assembly machine during camera positioning

Fig. 11. Absolute value of assembly positioning error