MULTIPLE SIZE REDUCTIONS FOR IMAGE SEGMENTATION

• Main Author: KRTOLICA RADOVAN V
• Format: Patent
• Language: English
• Published: 16.01.2003
• Edition: 7
• Subjects: ADVERTISING; ARRANGEMENTS OR CIRCUITS FOR CONTROL OF INDICATING DEVICESUSING STATIC MEANS TO PRESENT VARIABLE INFORMATION; CALCULATING; COMPUTING; COUNTING; CRYPTOGRAPHY; DISPLAY; EDUCATION; HANDLING RECORD CARRIERS; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; SEALS

A sample image (144) is segmented by an image segmentation system (120) including a size reduction unit (134), which reduces the size of the image (144), and, at the same time, fills small gaps between foreground pixels. Thereafter, a connected component analyzer (136) identifies connected components and their associated minimum bounding rectangles in the reduced image (145). Next, a target object filter (138) searches the connected components for target objects, making use of a target object library (146) to identify target objects characterized by such parameters as size, shape, and texture. Finally, an inverse mapper (140) locates the bounding rectangles of the target objects in the original sample image (144), and extracts the associated portions of the image (144) for analysis in a conventional image classifier (142).