Abstract:
An approach is provided for acquiring images with camera-enabled mobile devices using objects of interest recognition. A mobile device is configured to acquire an image represented by image data and process the image data to identify a plurality of candidate objects of interest in the image. The plurality of candidate objects of interest may be identified based upon a plurality of low level features or “cues” in the image data. Example cues include, without limitation, color contrast, edge density and superpixel straddling. A particular candidate object of interest is selected from the plurality of candidate objects of interest and a graphical symbol is displayed on a screen of the mobile device to identify the particular candidate object of interest. The particular candidate object of interest may be located anywhere on the image. Passive auto focusing is performed at the location of the particular candidate object of interest.
Abstract:
Disclosed is a system and method for obtaining ad-hoc the spatial layout and setup of installed radio-based indoor tracking systems at particular locations. The systems and methods can be implemented via an installed or web-based application that performs the ad-hoc detection of indoor beacons, which includes detecting the signal strength of all beacons at a location while the user performs a previously configured gesture. From the trajectory of the gesture and the recorded signal strengths, the disclosed systems and methods derive the distance and direction of all beacons, which are laid out in a 2D map. Thus, the present disclosure can implicitly calibrate signal strengths to high-accuracy distances, thereby producing the spatially accurate 2D setup of the previously unknown space, in addition to indoor tracking on a decimeter level, which is efficient and increasingly cost effective for device tracking and advertising purposes.
Abstract:
Disclosed is a system and method for calibrating BLE signal strengths to high-accuracy/precise distances. The present disclosure involves auto-calibrating BLE-based tracking systems, such as, for example, those used indoors using acoustic signals. The present disclosure enables BLE-based distance estimation to be accurate to decimeters and centimeters. The disclosed systems and methods utilize signals communicated to and from roaming devices in order to determine the distance(s) between the roaming device and installed BLE units. A signal-strength to distance map can then be constructed for reuse on any device with a Bluetooth component.