Abstract: This paper presents image segmentation and texture analysis algorithms on synthetic and real images. This research work demonstrates the considerable variability in an image understanding system performance based on different choices of image segmentation and texture analysis algorithms used. This research work includes results of a segmentation method to extract the object based on color and texture features of color images. Image segmentation denotes a process of partitioning an image into distinct regions. Based on the color segmentation result, and the texture variances between the background image and the object, we extract the object by the gray level co-occurrence matrix for texture segmentation. The GLCMs broadly represent the joint possibility of occurrence of grey-levels for pixels with a given spatial relationship in a defined region. Finally, the segmentation result is improved by mathematical morphology methods

• Publication Date: 02-Jun-2014
• DOI: 10.15224/978-1-63248-010-1-19
• Views: 0
• Downloads: 0

Abstract: This paper introduces a new trend for face recognition features. It is based on dividing the face into four horizontal and five vertical regions [16]. Each region is divided into an optimum number of eight vertical and seven horizontal partitions. One approach using average per partition features and another using histogram per region features are considered. Algorithms to find the minimum of Euclidean distance (ED) between a test image and a matching DB registered image are discussed. Both algorithms achieved 100% recognition rate (RR) with ORL and Yale databases. A new definition for RR that is termed inclusive recognition rate (RRi) is suggested. RRi supports testing images belonging to DB subject's images or not.

• Publication Date: 02-Jun-2014
• DOI: 10.15224/978-1-63248-010-1-20
• Views: 0
• Downloads: 0