Image resizing using saliency strength map and seam carving for white blood cell analysis

Area of Science:

• Medical Imaging
• Computer Vision
• Image Processing

Background:

• Conventional image resizing methods often distort or remove critical cellular structures.
• Preserving white blood cells in blood cell images is crucial for accurate medical diagnosis.

Purpose of the Study:

• To develop an advanced image resizing method that prioritizes the retention of important cellular components.
• To enhance the efficiency and quality of image resizing for medical applications.

Main Methods:

• A novel approach combining seam carving with a Saliency Strength Map (SSM) was developed.
• An energy map was created utilizing visual attention models and the structural characteristics of white blood cells.
• The energy map was optimized to maximize the significance of white blood cells while minimizing background and red blood cell prominence.

Main Results:

• The proposed method demonstrated superior image resizing capabilities compared to existing techniques.
• Quantitative analysis using Peak Signal-to-Noise Ratio (PSNR) and Ratio of Distortion (ROD) validated the method's effectiveness.
• Experimental results confirmed efficient size reduction while maintaining the integrity of white blood cells.

Conclusions:

• The Saliency Strength Map-based seam carving method significantly improves the preservation of critical structures in resized medical images.
• Future research aims to accelerate the resizing process for medical images without compromising visual fidelity.