Underwater Object Recognition Using Transformable Template Matching Based on Prior Knowledge

Figure 3

Construction process of the template based on prior knowledge: (a) is the previous frame, (b) is the next frame, (c) is the result of image difference and threshold segmentation, (d) is the erosion result of the candidate object, (e) is the dilation result of the candidate object, and (f) is the template containing the acoustic shadow and highlight region.

(a)

(b)

(c)

(d)

(e)

(f)