Systems and methods for classifying objects in digital images captured using mobile devices

作者: Jan Willers Amtrup , Anthony Macciola , Christopher W. Thrasher , Jiyong Ma , Stephen Michael Thompson

DOI:

关键词: Haar-like featuresRepresentation (systemics)Object-class detectionDigital image3D single-object recognitionFeature vectorComputer visionMethodFeature (computer vision)Computer scienceArtificial intelligence

摘要: In one embodiment, a method includes receiving digital image captured by mobile device; and using processor of the device: generating first representation image, being characterized reduced resolution; feature vector based on representation; comparing to plurality reference matrices; classifying an object depicted in as member particular class at least part comparing.