Abstract:
A system and method for generating and using a correlation filter. The method includes providing a plurality of training images, each training image being paired with an associated target correlation plane. Each training image and target correlation plane pair is processed. A final filter is generated, wherein the final filter is useable to generate a filtered output correlation plane of each training image. The final filter is selected to collectively minimize errors between the filtered output correlation plane of each training image and its associated target correlation plane. The final filter can be used in a wide variety of still image and video based object location and tracking applications.
Abstract:
A methodology is described to reduce the complexity of filters for face recognition by reducing the memory requirement to, for example, 2 bits/pixel in the frequency domain. Reduced-complexity correlations are achieved by having quantized MACE, UMACE, OTSDF, UOTSDF, MACH, and other filters, in conjunction with a quantized Fourier transform of the input image. This reduces complexity in comparison to the advanced correlation filters using full-phase correlation. However, the verification performance of the reduced complexity filters is comparable to that of full-complexity filters. A special case of using 4-phases to represent both the filter and training/test images in the Fourier domain leads to further reductions in the computational formulations. This also enables the storage and synthesis of filters in limited-memory and limited-computational power platforms such as PDAs, cell phones, etc. An online training algorithm implemented on a face verification system is described for synthesizing correlation filters to handle pose/scale variations. A way to perform efficient face localization is also discussed. Because of the rules governing abstracts, this abstract should not be used to construe the claims.
Abstract:
An apparatus and method for implementing a user interface module for use in screening a receptacle to detect therein the presence of one or more prohibited objects is provided. An image signal associated with the receptacle conveying information related to the receptacle's contents and a detection signal conveying a presence of at least one prohibited object in the receptacle are received. A user interface module is adapted for displaying first information conveying an image associated with the receptacle on the basis of the image signal. The user interface module is also adapted for displaying, simultaneously with the first information, second information conveying the presence of the prohibited object in the receptacle. The second information is derived at least in part on the basis of the detection signal. As a variant, the user interface module is adapted for providing a control allowing a user to cause third information to be displayed. The third information conveys at least one characteristic associated to the prohibited object. Alternative implementations of the user interface module may be made for use in screening a person to detect thereon the presence of one or more prohibited objects.
Abstract:
A method and apparatus for recognizing target objects. The method comprises identifying a group of target objects in an image. Further, the method comprises forming a group of chips encompassing the group of target objects identified in the image. Yet further, the method comprises recognizing the group of target objects in the group of chips using a group of filters, wherein the group of filters was created using a group of models for reference objects and environmental information for a location where the group of target objects was located when the image was generated, wherein a filter in the group of filters comprises a group of reference images for a reference object in the reference objects.
Abstract:
Disclosed is a distortion invariant system, method and computer readable medium for detecting the presence of one or more predefined targets in an input image. The input image and a synthetic discriminant function (SDF) reference image are correlated in a shift phase-encoded fringe-adjusted joint transform correlation (SPFJTC) correlator yielding a correlation output. A peak-to-clutter ratio (PCR) is determined for the correlation output and compared to a threshold value. A predefined target is present in the input image when the PCR is greater than or equal to the threshold value.
Abstract:
A signal processing device is provided, the device having a signal input for receiving a signal conveyed over a channel and defining a received signal. The device further includes one or more filters for generating a signal response based upon the received signal. The signal response includes an estimated value of a correntropy statistic. Additionally, the device includes a decision module connected to at least one of the filters for probabilistically deciding whether the received signal contains an information signal component based upon the estimated value of the correntropy statistic. The device further includes a signal output to convey a signal output indicating the received signal corresponds to a known signal template if the decision module decides that the received signal contains the information signal component.
Abstract:
A system for screening luggage items, which includes an image generation device suitable for generating an image signal associated with a luggage item where the image signal conveys information related to the contents of the luggage item. The system also includes an apparatus having an input for receiving the image signal and a processing unit. The processing unit processes the image signal in combination with a plurality of target images associated with target objects to detect a presence of at least one target object in the luggage item. The processing unit generates a detection signal in response to detection of the presence of at least one target object in the luggage item. An output module conveys information derived at least in part on the basis of the detection signal to a user of the system. In alternative embodiments, the system may also be used to screen cargo containers and persons.
Abstract:
A system and method for generating and using a correlation filter. The method includes providing a plurality of training images, each training image being paired with an associated target correlation plane. Each training image and target correlation plane pair is processed. A final filter is generated, wherein the final filter is useable to generate a filtered output correlation plane of each training image. The final filter is selected to collectively minimize errors between the filtered output correlation plane of each training image and its associated target correlation plane. The final filter can be used in a wide variety of still image and video based object location and tracking applications.
Abstract:
A system for screening cargo containers is provided including an image generation device, an apparatus for screening cargo containers and an output module. The image generation device generates an image signal conveying information related to the contents of the cargo container. The apparatus receives the image signal and a list of objects conveying objects expected to be present in the cargo container. A processing unit processes the image signal in combination with the list of objects and a group of target images associated with objects to derive mismatch information data. The mismatch information data conveys at least one distinction between the list of objects and the information related to the contents of the cargo container conveyed by the image signal. Information conveying the mismatch information data is released and conveyed to a user of the system by an output module.
Abstract:
Methods for using training images to define correlation strength surfaces about reference filters, permitting direct solution of position and orientation using images of a target object with only a few reference filters are disclosed. Correlations of selected reference filters with training images are used to develop expressions describing shapes of respective correlation surfaces at incremental changes relative to an aspect or range used for a filter. Correlations of target images with only a few reference filters may then be used to define target orientation and range in many real-world applications. A method for determining target object orientation and/or range based on interpolations, using results of correlations of a target image with composite filters having reduced sensitivities for different orientation axes, or range, is also disclosed. Techniques for using training images to create such composite filters are disclosed, as are extensions of methods to additional dimensions.