Robust Face Recognition via Multi-channel Models and Feature Selection
MetadataShow full item record
This thesis proposes various novel approaches to improve the performance of face recognition via using multi channels for face image representation. Methods proposed in this thesis are designed to solve different face recognition problems including Opened Mouth, Occlusions, Illumination variations, and Pose variations. All proposed methods have achieved significant improvements over state of the art methods.
Showing items related by title, author, creator and subject.
Dincer, Tuna (2000)Lactose is the major carbohydrate in milk. The presence of lactose in whey constitutes a significant pollution problem for dairy factories. At the same time, there is an increasing market for high quality crystalline ...
A randomised controlled trial of an online fatigue self-management group intervention for adults with chronic neurological conditionsGhahari, Setareh (2009)Background: Fatigue is one of the most common symptoms of neurological conditions. Although the literature suggests different approaches to treatment of this pervasive symptom, there is not a single, agreed comprehensive ...
Investigating the effectiveness of an online course : development of the comparative learning environment questionnaireIyer, Radha (2011)This study was undertaken with the purpose of evaluating a newly-developed online course. The study involved, firstly, designing, developing and validating two questionnaires that could be used to assess the relative ...