Abstract: A method of determining similarity of objects in images. Feature vectors are determined for objects in images captured by cameras operating in a training domain. Feature vectors are determined for the objects in images captured by cameras operating in a target domain, the cameras of the target domain operating with different environmental factors to the cameras of the training domain. A mapping is determined for a difference in the feature vectors of the training domain and the target domain. The difference in the feature vectors of the training domain and the target domain is converted to a matching space by applying the determined mapping to the feature vectors of the training domain and the target domain. A classifier is determined using data associated with the feature vectors of the training domain in the matching space.

Abstract: A method and system for enhancing spatial resolution of an image of an object includes using an image detector with a sensor array of pixels configured for extended dynamic range detection. Multiple image sample are collected to assemble a stack of image samples and input into an image processor for processing using an image reconstruction engine to generate a single output image from the stack. In a preferred embodiment, the image reconstruction engine executes the Pixon method for image reconstruction.

Abstract: In one embodiment a plurality of patches of an image are processed, using a first set of layers of a convolutional neural network, to output a plurality of object proposals associated with the plurality of patches of the image. Each patch includes one or more pixels of the image. Each object proposal includes a prediction as to a location of an object in the respective patch. Using a second set of layers of the convolutional neural network, the plurality of object proposals outputted by the first set of layers are processed to generate a plurality of refined object proposals. Each refined object proposal includes pixel-level information for the respective patch of the image. The first layer in the second set of layers of the convolutional neural network takes as input the plurality of object proposals outputted by the first set of layers.

Abstract: A method of comparing objects in images. A dictionary determined from a plurality of feature vectors formed from a test image and codes formed by applying the dictionary to the feature vectors is received, the dictionary being based on a difference in mean values between the codes. Comparison codes are determined for the objects in the images by applying the dictionary to feature vectors of the objects in the images. The objects in the images are compared based on the comparison codes of the objects.

Abstract: The present invention discloses a stepwise refinement detection method for pavement cracks, processing pavement images comprising the following primary steps: extracting and processing marking lines, extracting ROA, adaptive threshold segmentation based on ROA, extracting ROC and region growth based on ROC direction feature weighting.

Abstract: Provided is a disclosure for detection and recognition of objects using a projected array of light shapes of one or more frequencies, where the light may be structured or non-structured.

Abstract: An image capturing apparatus which generates an image by shooting a subject, calculates a parameter for image encoding, and generates encoded data by encoding the shot image using the parameter, wherein while encoded data of a first image to be encoded is generated, a first parameter is generated for encoding a second image to be encoded next based on an encoding result of a third image that has been encoded immediately before the first image, and in a period from completion of encoding of the first image to a start of encoding of the second image, one of the first parameter and a second parameter is provided as a parameter for encoding the second image based on a correlation between the first image and the third image.

Abstract: An image acquisition unit acquires first and second radiation images from first and second radiation detectors. A first frequency analysis unit generates a first high-frequency component and a first low-frequency component of the first radiation image, and a second frequency analysis unit generates a second low-frequency component of the second radiation image. A subtraction processing unit generates a subtraction low-frequency component from the first low-frequency component and the second low-frequency component, and a synthesis unit synthesizes the subtraction low-frequency component and the first high-frequency component to generate a processed image.

Abstract: Systems and methods for automatically determining of garment sizing (e.g., fit) using a images including video images. The method may perform non-contact estimations of garment fit from visual (e.g., video) input by receiving an video of the subject's head and face and profile and determining a scaling factor from the subject's intraocular spacing and using this scaling factor when analyzing images of other body regions to determine garment sizing.

Abstract: Architecture and a method for maximally stable extremal regions (MSERs)-based detection of exudates in an ocular fundus is disclosed. The architecture includes a communication interface to receive pixels of an ocular fundus image. The architecture further includes processing circuitry that is coupled to the communication interface. The processing circuitry is configured to automatically provide labels for light image regions and dark image regions within the ocular fundus image for a given intensity threshold and find MSERs within the ocular fundus image based on the labels. The architecture also determines MSER regions based on the MSER criteria and then highlights the pixels of the ocular fundus image that are located within MSER regions to indicate the exudates in the ocular fundus. The architecture is further configured to determine MSER ellipses parameters based on MSER regions and MSER criteria and then highlight the locations of the exudates in the ocular fundus.

Abstract: A method determining a retinal pigment epithelium identifies a plurality of regions using captured image data of the eye and fits a curve into at least some of the regions. A curve score is determined associated with the fitted curve using at least a distance between the fitted curve and at least some of the regions in which a contribution of the regions to the curve score is biased towards (asymmetric) regions below the fitted curve. These steps are repeated whereupon one of the fitted curves is selected, using the corresponding associated curve score, for classifying some of the regions as forming at least a part of a retinal pigment epithelium.

Abstract: An image processing device includes: a memory; and a processor coupled to the memory and the processor configured to execute a process, the process comprising: acquiring a plurality of biometric images of a same biological body part of a same user from a biometric sensor; generating, when a feature point appears at a corresponding position of the same biological body part in one or more of the plurality of biometric images and a feature point does not appear at the corresponding position in another of the plurality of biometric images, a pseudo feature point at the corresponding position of the another; and storing information of the feature point and the pseudo feature point of the plurality of biometric images as a biometric feature.

Abstract: The present invention discloses methods, systems, and computer programmable products for tracking objects across a first frame and a second frame of a video. An object tracking system computes a cost function between each object in the first frame and each object in the second frame. Further, a pair of objects is selected from the first frame and the second frame based on a pre-defined criteria which is based on the computed cost function. Thereafter, the object tracking system established a correspondence between the selected pair of objects.

Abstract: The invention, in one aspect, is related to a method of determining spatial coordinates of a 3D reconstruction of at least part of a first real object at absolute spatial scale comprising the steps of: receiving image information of a first image including at least part of the first real object captured with a first camera, of a third image including the at least part of the first real object captured with a third camera, of a second image including at least a first part of a second real object captured with a second camera, and of a fourth image including at least a second part of the second real object captured with a fourth camera, wherein the frustum of the first camera and the frustum of the second camera do not overlap and wherein the frustum of the third camera and the frustum of the fourth camera do not overlap, and providing a first spatial transformation between the first camera and the second camera and providing a second spatial transformation between the third camera and the fourth camera, provid

Abstract: A system for preliminary diagnosis of ocular diseases is proposed, in which from capturing a plurality of images of the eyes of a person, a final image, corrected by processing the images by a computer application, is obtained. The already mentioned application calculates the percentage of the colors composing the pupillary reflex of each eye, and compares the results with previous reference pictures obtained from normal cases and clinical cases of ocular diseases.

Abstract: In one embodiment, the apparatus includes a decoder. The decoder is configured to obtain first and second motion vectors of a block other than the current block. The other block neighbors the current block at one of a left, top and top right position. The decoder is configured to determine first and second motion vectors of the current block using the first and second motion vectors of the other block such that the first motion vector of the current block has a same direction as the first motion vector of the other block and the second motion vector of the current block has a same direction as the second motion vector of the other block.

Abstract: Methods, devices and computer-readable mediums for detecting occlusions which occur due to foreground object movement with respect to a background between first and second successive frames. Occlusion detection may use motion estimation with respect to at least a third frame temporally preceding the first frame. Occlusion detection may be based on one or more assumptions such as: occlusion motion vectors are different than other background motion vectors; occlusion motions are likely to be similar to foreground occluding motion; and/or motion estimation will match an occlusion block with a block belonging to a common background object. Occlusion detection may be combined with motion error detection based on a motion field divergence using a motion vector assigned to the occlusion, e.g., for generating an intermediate frame in frame up rate conversion (FRUC).

Abstract: A procedure can be assisted by a processor system, such as a computer system. A trajectory can be used to identify a selected trajectory or path of an instrument to reach a tumor within a brain of a subject, reach a selected portion of the anatomy (e.g. sub-thalamic nucleus (STN) or spinal cord), or other appropriate target. The planning algorithm can include both inputted data and learned rankings or ratings related to selected trajectories. The planning algorithm can used the learned ratings to rate and later determined trajectories.

Abstract: A robust segmentation technique based on multi-layer classification technique to identify the lesion boundary is described. The inventors have discovered a technique based on training several classifiers such that to classify each pixel as lesion versus normal Each classifier is trained on a specific range of image resolutions. Then, for a new test image, the trained classifiers are applied on the image. Then by fusing the prediction results in pixel level a probability map is generated. In the next step, a thresholding method is applied to convert the probability map to a binary mask, which determines a mole border.

Abstract: An inspection apparatus, system, and method, each of which: acquires a master image serving as an inspection reference, the master image being generated based on image data to be printed as a printed image; acquires a read image read from the printed image; extracts a neighboring region neighboring an edge region of the master image; determines whether a change amount of density of pixels in the neighboring region falls within a predetermined range; based on a determination that the change amount of density of pixels in the neighboring region falls within the predetermined range, calculates a statistic of density of pixels in a corresponding region of the read image, which corresponds to the neighboring region of the master image; and determines existence or non-existence of a defect in the corresponding region of the read image based on the statistic of the corresponding region.