New fusion operations for digitized binary images and their applications

Area of Science:

• Computer Vision
• Digital Image Processing
• Computational Geometry

Background:

• Binary image processing relies on fundamental operations like expansion and contraction.
• Classical fusion operations often involve sequential application of expansion and contraction.
• Existing methods may have limitations in preserving image topology or handling complex structures.

Purpose of the Study:

• To propose a new class of fusion operations for binary image processing.
• To introduce algorithms for these novel fusion operations.
• To demonstrate the effectiveness of these operations in various image processing tasks.

Main Methods:

• Development of an algorithm for classical fusion operations using distance transformation.
• Definition of topological contraction and expansion preserving the Euler number.
• Sequential combination of classical and topological operations to create new fusion methods.

Main Results:

• The proposed algorithms effectively implement classical and novel fusion operations.
• Topological contraction erases components while preserving the Euler number.
• Topological expansion fills holes while preserving the Euler number.
• Experimental results validate the utility in shape smoothing, component merging/separation, and structure analysis.

Conclusions:

• The new fusion operations offer enhanced capabilities for binary image processing.
• Combinations of classical and topological operations provide versatile tools.
• These methods are effective for tasks including shape analysis and component manipulation.