Adaptively Active Contours Based on Variable Exponent Lp(|∇I|) Norm for Image Segmentation

Figure 1

Results of our model with 𝑝(|∇𝐼|)=2 and 𝑝(|∇𝐼|)>2 for three infrared images with noise (the first two images) and low contrast (the last image), (left to right: 𝜇=0.09×255×255,0.001×255×255,0.1×255×255), where initial and final contours are shown in green and red, respectively. Top row: 𝑝(|∇𝐼|)=2, at the 75, 100, 50 iterations. Bottom row: 𝑝(|∇𝐼|)>2 (𝐾=5.2,4.1,6.4), at the 1st iteration.

(a)

(b)

(c)

(d)

(e)

(f)