Abstract: A system and method that performs iterative foreground detection and multi-object segmentation in an image is disclosed herein. A new background prior is introduced to improve the foreground segmentation results. Three complimentary methods detect and segment foregrounds containing multiple objects. The first method performs an iterative segmentation of the image to pull out the salient objects in the image. In a second method, a higher dimensional embedding of the image graph is used to estimate the saliency score and extract multiple salient objects. A third method uses a metric to automatically pick the number of eigenvectors to consider in an alternative method to iteratively compute the image saliency map. Experimental results show that these methods succeed in accurately extracting multiple foreground objects from an image.

Abstract: A system and method that performs iterative foreground detection and multi-object segmentation in an image is disclosed herein. A new background prior is introduced to improve the foreground segmentation results. Three complimentary methods detect and segment foregrounds containing multiple objects. The first method performs an iterative segmentation of the image to pull out the salient objects in the image. In a second method, a higher dimensional embedding of the image graph is used to estimate the saliency score and extract multiple salient objects. A third method uses a metric to automatically pick the number of eigenvectors to consider in an alternative method to iteratively compute the image saliency map. Experimental results show that these methods succeed in accurately extracting multiple foreground objects from an image.

Abstract: A system and method that performs iterative foreground detection and multi-object segmentation in an image is disclosed herein. A new background prior is introduced to improve the foreground segmentation results. Three complimentary methods detect and segment foregrounds containing multiple objects. The first method performs an iterative segmentation of the image to pull out the salient objects in the image. In a second method, a higher dimensional embedding of the image graph is used to estimate the saliency score and extract multiple salient objects. A third method uses a metric to automatically pick the number of eigenvectors to consider in an alternative method to iteratively compute the image saliency map. Experimental results show that these methods succeed in accurately extracting multiple foreground objects from an image.

Abstract: A system and method that performs iterative foreground detection and multi-object segmentation in an image is disclosed herein. A new background prior is introduced to improve the foreground segmentation results. Three complimentary methods detect and segment foregrounds containing multiple objects. The first method performs an iterative segmentation of the image to pull out the salient objects in the image. In a second method, a higher dimensional embedding of the image graph is used to estimate the saliency score and extract multiple salient objects. A third method uses a metric to automatically pick the number of eigenvectors to consider in an alternative method to iteratively compute the image saliency map. Experimental results show that these methods succeed in accurately extracting multiple foreground objects from an image.

Abstract: A joint torque augmentation system includes linkage assembly configured to couple to a user. Linkage assembly includes a unidirectional link and a device joint. The linkage assembly is worn by a user or is configured to couple to footwear. An actuator is coupled to the linkage assembly to provide a torque at a joint of the user. A sensor is coupled to the user to measure a position of the user. A control system is coupled to the sensor and actuator. A phase of gait for the user is determined by the control system based on the position measured by the sensor. The actuator produces a tension force on the linkage assembly during a first phase of gait. A compliant element is coupled between the actuator and linkage assembly. The compliant element is tuned based on a load carried by the user.

Abstract: A method of controlling a gait device includes a first sensor disposed on a first mobile body. A physical state of the first mobile body is measured using the first sensor to obtain a first physical state measurement. A second sensor is disposed on a second mobile body. A physical state of the second mobile body is measured using the second sensor to obtain a second physical state measurement. The first and second physical state measurements are conditioned by pseudo integration. A reference function is based on a gait activity. A reference function is determined by measuring a physical state of an able-bodied human and correlating an output position of the actuator to the physical state of the able-bodied human. A command is generated by inputting the first and second physical state measurements into the reference function to control an actuator of the gait device to match the output position.

Abstract: A load support device has a first link assembly coupled to a load and a first foot of a user. A first damping element is coupled to the first link assembly. The first damping element includes a double acting piston configured to provide uni-directional damping. A first sensor is disposed on a first limb of the user. A physical characteristic of the first limb is measured with the first sensor. A damping constant of the first damping element is selected based on the physical characteristic of the first limb. A second link assembly is coupled to the load and to a second foot of the user. A second damping element is coupled to the second link assembly between the load and the second foot. The load is alternately supported by the first link assembly and damping element and the second link assembly and damping element throughout a gait cycle.

Abstract: A joint torque augmentation system includes linkage assembly configured to couple to a user. Linkage assembly includes a unidirectional link and a device joint. The linkage assembly is worn by a user or is configured to couple to footwear. An actuator is coupled to the linkage assembly to provide a torque at a joint of the user. A sensor is coupled to the user to measure a position of the user. A control system is coupled to the sensor and actuator. A phase of gait for the user is determined by the control system based on the position measured by the sensor. The actuator produces a tension force on the linkage assembly during a first phase of gait. A compliant element is coupled between the actuator and linkage assembly. The compliant element is tuned based on a load carried by the user.

Abstract: A method of controlling a gait device includes a first sensor disposed on a first mobile body. A physical state of the first mobile body is measured using the first sensor to obtain a first physical state measurement. A second sensor is disposed on a second mobile body. A physical state of the second mobile body is measured using the second sensor to obtain a second physical state measurement. The first and second physical state measurements are conditioned by pseudo integration. A reference function is based on a gait activity. A reference function is determined by measuring a physical state of an able-bodied human and correlating an output position of the actuator to the physical state of the able-bodied human. A command is generated by inputting the first and second physical state measurements into the reference function to control an actuator of the gait device to match the output position.

Abstract: A method of adjusting exposure of a digital camera based on range information, including a digital camera capturing a first digital image at a selected exposure of a scene having objects; providing range information having two or more range values indicating the distance from the digital camera to objects in the scene; using the range information and pixel values of the captured digital image to determine an exposure adjustment amount for the selected exposure; and applying the exposure adjustment amount to the digital image to produce a second digital image with adjusted exposure.

Abstract: A method of processing a digital image of a scene to locate a planar surface in the digital image, includes providing a range information including two or more range values indicating the distance of objects in the scene from a known reference frame; and using a planar model and the range information to locate one or more planar surfaces in the digital image.

Abstract: A method of detecting an object of interest having a known size in a digital image, includes providing a range information including two or more range values indicating the distance of objects in the scene from a known reference frame; detecting a candidate object of interest in the image; determining range values corresponding to the candidate object of interest and using these range values and the known size of the object of interest to classify the candidate object of interest.

Abstract: A method for detecting undesirable images within an image sequence, comprising the steps of receiving the image sequence; computing one or more associated image characteristics for a group of images, wherein the group of images includes at least one image from the image sequence; identifying any undesirable group of images based on the computed image characteristics; collecting undesirable groups of images into at least one collection; receiving a threshold value of a minimum number of undesirable images; and flagging images that belong to one or more collections that have at least the threshold value of undesirable images.

Abstract: A system for the detection of abnormalities in a medical image of a subject. The system includes an examination bundle, a learning engine, and a detecting engine. The examination bundle includes at least one medical image of the subject from a first modality and at least one medical image of the subject from a second modality. The learning engine is employed for determining the characteristics of abnormalities within the at least one medical image from the first modality and the at least one medical image from the second modality. The detecting engine is employed for detecting abnormalities within at least one of the medical images comprising the examination bundle.

Abstract: An image indexer for indexing a plurality of images that includes a first data structure for subsequent classification of the one or more images. The first data structure includes characteristics for at least one class. An image classifier classifies one or more individual images found in the plurality of images as classified images according to the first data structure. A second data structure performs subsequent clustering of the plurality of images, wherein the second data structure includes at least two sequential events in a set of known events. The classified images are clustered according to the second data structure, and a representative image is selected from each cluster of classified images.

Abstract: A method for delivering radiation therapy to a patient using a three-dimensional planning image for radiation therapy of the patient wherein the planning image includes a radiation therapy target. The method includes the steps of: determining desired image capture conditions for the capture of at least one two-dimensional radiographic image of the radiation therapy target using the three-dimensional planning image; detecting a position of the radiation therapy target in the at least one captured two-dimensional radiographic image; and determining a delivery of the radiation therapy in response to the radiation therapy target's detected position in the at least one captured two-dimensional radiographic image.

Abstract: A method for reconstructing computerized tomographic (CT) images of an object, including: scanning the object with a CT imaging system to acquire views that include measured projections of the object. Additionally, the method applies an iterative algorithm to minimize errors between the measured projections and reprojections of a reconstructed CT image, wherein at each iteration, projection errors become smaller causing the reconstructed CT image to become further refined.

Abstract: A system for identifying anatomical structure depicted in an in vivo image, that includes an examination bundlette having a captured in vivo image; and a gastrointestinal atlas that includes a list of individual anatomical structures and characterization data of the individual anatomical structures. A classification engine analyzes the examination bundlette and the gastrointestinal atlas to identify the anatomical structure depicted in the captured in vivo image.

Abstract: A digital image processing method for exposure adjustment of in vivo images that includes the steps of acquiring in vivo images; detecting any crease feature found in the in vivo images; preserving the detected crease feature; and adjusting exposure of the in vivo images with the detected crease feature preserved.

Abstract: A digital image processing method for real-time automatic abnormality notification of in vivo images and remote access of in vivo imaging system, comprising the steps of: acquiring multiple sets of images using multiple in vivo video camera systems; for each in vivo video camera system forming an in vivo video camera system examination bundlette; transmitting the examination bundlette to proximal in vitro computing device(s); processing the transmitted examination bundlette; automatically identifying abnormalities in the transmitted examination bundlette; setting off alarming signals locally provided that suspected abnormalities have been identified; receiving one or more unscheduled alarming messages from one or more endoscopic imaging systems randomly located; routing alarming messages to remote recipient(s); and executing one or more corresponding tasks in relation to the alarming messages.