Efficient Multi-Gait Recognition via Fast RCNN and Shadow Removal: Efficient Multi-gait Recognition Using Effective Shadow Removal And Classification Technique Using Fast RCNN - Couverture souple

M., Amanulla Khan; S., Selvaperumal; S., Syed Musthafa Gani

Couverture souple

ISBN 10 : 6209900070 ISBN 13 : 9786209900075

Editeur : LAP LAMBERT Academic Publishing, 2026

Synopsis

This book presents an efficient multi-gait recognition system using shadow removal and Fast RCNN-based classification techniques. Gait recognition identifies individuals through walking patterns, but shadows, illumination changes, and clothing variations reduce accuracy. To overcome these issues, a Deep Learning-based SDR-RAN model is proposed for shadow detection, removal, and image inpainting. The model uses GhostNet, batch normalization, and Parametric ReLU for effective feature extraction. CRANN restores shadow regions using background information. In addition, a Faster RCNN framework with RPN and fuzzy K-means clustering is developed for accurate gait classification. Experimental results show improved recognition performance for surveillance applications.

Détails de l'édition

• Éditeur: LAP LAMBERT Academic Publishing
• Date d'édition: 2026
• Langue: anglais
• ISBN 10: 6209900070
• ISBN 13: 9786209900075
• Reliure: Broché
• Nombre de pages: 140
• Coordonnées du fabricant: non disponible
• Personne responsable: non disponible