A direct approach for object detection with catadioptric omnidirectional cameras
MetadataShow full item record
In this paper, we present an omnidirectional vision-based method for object detection. We first adopt the conventional camera approach that uses sliding windows and histogram of oriented gradients (HOG) features. Then, we describe how the feature extraction step of the conventional approach should be modified for a theoretically correct and effective use in omnidirectional cameras. Main steps are modification of gradient magnitudes using Riemannian metric and conversion of gradient orientations to form an omnidirectional sliding window. In this way, we perform object detection directly on the omnidirectional images without converting them to panoramic or perspective images. Our experiments, with synthetic and real images, compare the proposed approach with regular (unmodified) HOG computation on both omnidirectional and panoramic images. Results show that the proposed approach should be preferred.
Showing items related by title, author, creator and subject.
Constant False Alarm Rate (CFAR) detection based estimators with applications to sparse wireless channels Karaca, Ümit (İzmir Institute of Technology, 2006)We provide Constant False Alarm Rate (CFAR) based thresholding methods for training based channel impulse response (CIR) estimation algorithms for communication systems which utilize a periodically transmitted training ...
A BODIPY-based reactive probe for the detection of Au(iii) species and its application to cell imaging A BODIPY-based fluorescent probe bearing a pyridyl hydrazone motif responds selectively to Au(III) ions through an irreversible C[double bond, length as m-dash]N bond hydrolysis reaction. Gold species, besides their ...
Tarhan, Çiğdem (İzmir Institute of Technology, 2007)This thesis is an example of how land use changes could be detected via high resolution remotely sensed data in GIS environment. In order to perform .change detection. IKONOS satellite images, belonging to 2001 and 2004, ...