Available parking slot recognition based on slot context analysis

• DOI: 10.1049/iet-its.2015.0226
• Publication Date: 2016-08-15T11:09:20Z
• Authors: Soomok Lee; Seung‐Woo Seo
• Keywords: 0202 electrical engineering, electronic engineering, information engineering; 02 engineering and technology

Vacant parking‐slot recognition is one of the core elements for achieving a fully automated parking assistant system. The vacant typically executed in two stages: and slot‐occupancy classification. In most previous works, considering recognised slot results as grid cell, existence objects evaluated through examination Despite moderate performance, however, these methods cannot distinguish availability moving or small permit marks, since all only focus on occupancy parked vehicles. this study, authors present camera‐based available algorithm based slot‐context analysis. By utilising abundant visual features, they propose new handling diverse conditions. For recognition, methodology extracting associating line‐markings. proposed includes slot‐validation step that identifies multiple contexts probabilistic integration has more flexibility irregular patterns. classification stage, reliable parking‐availability detection achieved visual‐slot features including histogram gradient frequency–magnitude via support vector machine. Simulations vehicle‐level experiments demonstrate robustness