More recently, the smartphone intergrated powerful camera is an efficient platform for location-wareness. The matching of smartphone recordings with a database of geo-referenced images allows for meter accurate infrastructure-free localization. However, for high accuracy indoor positioning
using a smartphone, there are two constraints that includes: (1) limited computational and memory resources of smartphone; (2) user’s moving in large buildings. These constraints are also typically more severe for systems that should be wearable and used indoors. To address these issues,
we proppose a novel smartphone camera-based algorithm for supporting a scalability and high accuracy indoor positiong service. In order to obtain an accurate image matching, we proppose a new feature descriptor that efficiently fused of HOG and LPQ feature. The novel feature is the local phase
quantization of a salient HOG visualuizing image. The specific properties of this feature is robust in the indoor scenarios. In order to reduce the network latency and communications traffic, we introduce a basestation based indoor positiioning system for providing a coarse location. Comparing
to other states of art methods, experimental results show that our algorithm allowed instantaneous camera-based indoor positioning with very low requirements on the available network connection.