A genetically modified fuzzy linear discriminant analysis for face recognition

This paper addresses the face recognition problem through a modification of the Fuzzy Fisherface classification method. In conventional methods, the relationship of each face to a class is assumed to be crisp. The Fuzzy Fisherface method introduces a gradual level of assignment of each face pattern to a class, using a membership grading based upon the K-Nearest Neighbor (KNN) algorithm. This method was further modified by incorporating the membership grade of each face pattern into the calculation of the between-class and within-class scatter matrices, termed as Complete Fuzzy LDA (CFLDA). The present work aims at improving the assignment of class membership by improving the parameters of the membership functions. A genetic algorithm is employed to optimize these parameters by searching the parameter space. Furthermore, the genetic algorithm is used to find the optimal number of nearest neighbors to be considered during the training phase. The experiments were performed on the Olivetti Research Laboratory (ORL) face image database and the results show consistent improvement in the recognition rate when compared to the results from other techniques applied on the same database and reported in literature.     

Assessing multiple object tracking in young children using a game

Visual tracking of multiple objects in a complex scene is a critical survival skill. When we attempt to safely cross a busy street, follow a ball’s position during a sporting event, or monitor children in a busy playground, we rely on our brain’s capacity to selectively attend to and track the position of specific objects in a dynamic scene. This ability to visually track simultaneously moving objects in a continuously changing and multisensory environment is a critical component of nearly all forms of visual-motor coordination. While methods for assessing Multiple Object Tracking (MOT) in adults are well established, due to challenges associated with designing a MOT task suitable for young children, we have little understanding of MOT abilities under the age of 5 years. To better understand how and when young children learn to track multiple objects, we designed, implemented and evaluated TrackFX, the first game-based MOT task running on a touch tablet designed for children as young as 30 months old. We present findings from an empirical study of 31 children between the age of 30 and 58 months and implications for game-based learning.