A Novel Gradient Vector Flow Snake Model Based on Convex Function for Infrared Image Segmentation.

Area of Science:

• Computer Vision
• Image Processing
• Artificial Intelligence

Background:

• Infrared image segmentation is difficult due to noise, low contrast, and weak edges.
• Active contour models, like Gradient Vector Flow (GVF), are useful but struggle with noise smoothing versus edge protection.
• Existing GVF models significantly reduce segmentation effectiveness in infrared imagery.

Purpose of the Study:

• To propose a novel generalized gradient vector flow snakes model for enhanced infrared image segmentation.
• To address the limitations of traditional GVF models in balancing noise reduction and weak edge preservation.
• To improve the overall accuracy and robustness of infrared image segmentation.

Main Methods:

• Developed a novel generalized gradient vector flow (GGVF) snakes model by integrating Generic Gradient Vector Flow (GGVF) and Normally Biased Gradient Vector Flow (NBGVF) concepts.
• Implemented a new coefficient setting strategy using convex functions to enhance weak edge protection during noise smoothing.
• Conducted comparative experiments against existing snakes models using infrared image datasets.

Main Results:

• The proposed GGVF snakes model demonstrated superior performance in infrared image segmentation.
• The novel coefficient setting effectively protected weak edges while simultaneously smoothing noise.
• Experimental results showed significant improvements over traditional GVF and other snakes models.

Conclusions:

• The proposed generalized gradient vector flow snakes model offers a significant advancement for infrared image segmentation.
• The integration of GGVF and NBGVF, coupled with convex function-based coefficients, effectively overcomes the limitations of prior methods.
• This approach provides a more robust and accurate solution for segmenting challenging infrared images.