Fig. 1. Main flow chart of dynamic spectral coding image fusion

Fig. 2. Noise standard deviation of weight distribution image fusion with different spectral non-linear amplification noises

Fig. 3. (a) Schematic of three-primary-color receiving arrays; (b) spectral response curves

Fig. 4. Calibrated image of surface damage of the sample to be tested

Fig. 5. 20× magnified surface damage images. (a) Red light; (b) green light; (c) blue light; (d) white light

Fig. 6. 100× magnified images of single damage point. (a) Red light; (b) green light; (c) blue light; (d) white light

Fig. 7. Traditional edge feature binarization images. (a) Red light; (b) green light; (c) blue light; (d) white light

Fig. 8. Dynamic spectral coding image fusion effects. (a) Red light; (b) green light; (c) blue light; (d) average weight; (e) optimized fusion image

Fig. 9. Comparison of surface defects imaging and image fusion. (a) Red light; (b) green light; (c) blue light; (d) optimized fusion image