Abstract: Method and device for deblurring an out-of-focus blurred image: first, a preset blur kernel is used to carry out blurring processing on an original image to obtain a re-blurred image. Blur amounts of pixels in an edge area of the original image are estimated according to the change of the image edge information in the blurring processing to obtain a sparse blur amount diagram. Blur amounts of pixels in a non-edge area of the original image are estimated according to the sparse blur amount diagram to obtain a complete blur amount diagram. Deblurring processing is carried out according to the complete blur amount diagram to obtain a deblurred image. In the method and device provided, since a blur amount diagram is obtained based on the change of edge information after image blurring, the blur amount diagram can be more accurate, so that the quality of a deblurred image is improved.

Abstract: For example, a captured image including an image to be processed is received. In the image to be processed, various items are separately displayed in a plurality of areas. Subsequently, a partition pattern that partitions the image to be processed, which is included in the received captured image, into a plurality of areas is extracted. An object to be processed is recognized by comparing the extracted partition pattern with one or more partition patterns prestored in a storage unit and then identifying a partition pattern corresponding to the partition pattern extracted from the image to be processed. This makes it possible to recognize an object to be processed under a relatively light processing load.

Abstract: The present invention relates to a surveillance system and, more particularly, to a graphical display and user interface system that provides high-speed triage of potential items of interest in imagery. The system receives at least one image of a scene from a sensor. The image is pre-processed to identify a plurality of potential objects of interest (OI) in the image. The potential OI are presented to the user as a series of chips on a threat chip display (TCD), where each chip is a region extracted from the image that corresponds to a potential OI. Finally, the system allows the user to designate, via the TCD, any one of the chips as an actual OI.

Abstract: A Bayer-mask image is decoded by forming a decoded green array; calculating a slope at each pixel of the array, expressed as an angle; calculating an activity at each pixel; converting the slope angle and the activity into a complex number for each pixel, of modulus equal to the activity and argument equal to twice the slope angle; expressing said complex numbers in Cartesian coordinates to form a Cartesian slope signal and filtering the Cartesian slope signal with a linear spatial filter to derive a slope measure. Blue-green and red-green values are calculated and interpolated using a slope-adaptive interpolation filter steered by said slope measure.

Abstract: An inspection condition determination device comprises: an addition unit that adds data mimicking a flaw assumed to occur in an inspection target to a designated position of a three-dimensional model of the inspection target; a generation unit that generates an image without a flaw by replicating an optical condition for capturing an image of the inspection target on the three-dimensional model, and an image with the flaw by replicating the optical condition on the three-dimensional model to which the data mimicking the flaw is added; a determination unit that determines whether or not a difference between the image without a flaw and the image with the flaw at the designated position exceeds a threshold that allows detection of the flaw in the inspection target; and an extraction unit that extracts an optical condition available for detecting flaws of multiple designated patterns from multiple optical conditions.

Abstract: Methods, apparatus, and systems are provided for integrated driver expression recognition and vehicle interior environment classification to detect driver condition for safety. A method includes obtaining an image of a driver of a vehicle and an image of an interior environment of the vehicle. Using a machine learning method, the images are processed to classify a condition of the driver and of the interior environment of the vehicle. The machine learning method includes general convolutional neural network (CNN) and CNN with adaptive filters. The adaptive filters are determined based on influence of filters. The classification results are combined and compared with predetermined thresholds to determine if a decision can be made based on existing information. Additional information is requested by self-motivated learning if a decision cannot be made, and safety is determined based on the combined classification results. A warning is provided to the driver based on the safety determination.

Abstract: A method and system for segmenting a plurality of images. The method comprises the steps of segmenting the image through a novel clustering technique that is, generating a composite depth map including temporally stable segments of the image as well as segments in subsequent images that have changed. These changes may be determined by determining one or more differences between the temporally stable depth map and segments included in one or more subsequent frames. Thereafter, the portions of the one or more subsequent frames that include segments including changes from their corresponding segments in the temporally stable depth map are processed and are combined with the segments from the temporally stable depth map to compute their associated disparities in one or more subsequent frames. The images may include a pair of stereo images acquired through a stereo camera system at a substantially similar time.

Abstract: An object recognition device according to an embodiment includes a camera that captures an image of an imaging area. A storage device stores, for each of a plurality of registered objects, dictionary feature information for identifying the corresponding object and dictionary boundary information for identifying an actual boundary area of the corresponding object. A processor receives the captured image from the camera, and determines an object area in the captured image. The processor extracts feature information from the object area, and, based on the extracted feature information compared to the dictionary feature information, identifies each object included in the object area. The processor also extracts boundary information corresponding to each identified object included in the object area, and, based on the extracted boundary information compared to the dictionary boundary information with respect to each identified object, determines an overlap state of each identified object in the object area.

Abstract: A method for training a feature detector of an image processing device, including the steps of detecting features in the image to generate a score map, computing a center of mass on the score map to generate a location, extracting a patch from the image at the location by a first spatial transformer, estimating an orientation of the patch, rotating the patch in accordance with the patch orientation with a second spatial transformer, and describing the rotated patch to create a description vector.

Abstract: Methods, systems, and apparatus for providing indications of occluded objects. In some aspects a method includes the actions of determining a position of an observer whose view of an object is obstructed by a barrier; determining a position of the object relative to the observer; generating an indication of the position of the object based at least in part on the determined position of the observer; and displaying the indication of the position of the object on a display located between the observer and the barrier.

Abstract: In some examples, a method for compressing a spectral reflectance dataset may be performed through compression circuitry. The method may include computing a principal component analysis basis for the spectral reflectance dataset; projecting the spectral reflectance dataset onto the principal component analysis basis to obtain a weight matrix; quantizing the weight matrix; performing a Huffman encoding process on the quantized weight matrix to generate a Huffman table and Huffman codes for the quantized weight matrix; and providing compressed spectral reflectance data as the principal component analysis basis, the Huffman table, and the Huffman codes.

Abstract: Described is a system for trajectory estimation of a mobile platform, such as a UAV. In operation, the system generates an initial trajectory estimate for the mobile platform which is stored in a trajectory buffer as a buffered trajectory. Images captured at a location are compared with a location recognition database to generate a location label for a current location to designate the current location as a new location or a revisited location. If the location is a revisited location, the system determines if trajectory correction is required. If so, the buffered trajectory is corrected to generate a corrected trajectory as the drift-free trajectory. Finally, the drift-free trajectory can be used in a variety of applications. For example, the drift-free trajectory can be used to cause the mobile platform to traverse a path that coincides with the drift-free trajectory.

Abstract: An image retrieval apparatus includes a processor, and the processor performs a process including: determining an image in which a first characteristic object is included in a subject to be a first image, and determining an image that is captured after the first image and in which a second characteristic object is included in the subject to be a second image, from among a series of captured images; specifying images as an image group, the images being captured during a period after the first image is captured before the second image is captured from among the series of captured images; and extracting a representative image from the image group.

Abstract: A computer system running image processing software receives an identification of a desired scene of a geographical area for which an oblique-mosaic image is desired including one or more geometry parameters of a virtual camera; creates a mathematical model of the virtual camera having mathematical values that define the camera geometry parameters that configure the model to capture the geographical area, and looking down at an oblique angle; creates a ground elevation model of the ground and vertical structures within the oblique-mosaic pixel map, wherein source images were captured at an oblique angle and compass direction similar to the oblique angle and compass direction of the virtual camera; and reprojects, with the mathematical model, source oblique image pixels of the overlapping source images for pixels included in the oblique-mosaic pixel map using the ground elevation model to thereby create the oblique-mosaic image of the geographical area.

Abstract: A method and apparatus for measuring a distance to an object, by a device having at least two cameras, is described. One or more first images including the object are acquired by a first camera of the device and one or more first reference images are acquired by a second camera of the device, while the device is in a first position. One or more second images including the object and one or more second reference images are acquired by cameras of the device, while the device is in a second position, different from the first position. Based on the first and second reference images, information on the displacement of at least one camera of the device between the first and second position is determined. The distance from the device to the object is calculated based on the first and second images and the determined information on the displacement.

Abstract: A method for inspecting engravings in a tire mold includes inputting in advance into a data processing device two-dimensional image master data of an engraving formed on a surface of a side plate in a plan view using basic data set as a reference, the image master data having a radial component and a radial position of the engraving corrected on the basis of a locus length in a radial direction of the surface of the side plate; generating two-dimensional image actual measurement data, the image actual measurement data having a radial component and a radial position of three-dimensional data acquired via actual measurement corrected on the basis of a radial locus length of a surface of a side plate on which actual measurement was performed; and determining whether or not an image shape and an image position of the image actual measurement data is within a tolerable range.

Abstract: Currency bills are transported past an image scanner to one or more output receptacles. Each of the bills is imaged to produce image data from which a visually readable image of each bill can be reproduced. The serial number, denomination, and/or secondary identifiers of a bill is attempted to be extracted and/or determined from the image data associated with the bill. The serial number of the bill has an integer number, X, of characters. One or more of the X characters of the serial number of the currency bill is not extracted with a predetermined confidence. In response to failing to extract all of the X characters of the serial number of the bill with the predetermined confidence, a serial number field in an electronic record associated with the bill is populated with a serial number snippet image. The electronic record is stored in a non-transitory memory.

Abstract: A method and control system for optimizing active measurements are presented. In an embodiment, the method includes emitting at least one active signal in at least one direction; determining whether a return signal is detected for each of the at least one active signal; upon detecting at least one return signal for the at least one active signal: measuring a strength of each return signal; comparing the measured strength of each return signal to at least one optimal return signal strength; and determining, based on the comparison, an optimized configuration for active measurement emissions in one or more directions.

Abstract: A method for learning parameters of a CNN capable of detecting ROIs determined based on bottom lines of nearest obstacles in an input image is provided. The method includes steps of: a learning device instructing a first to an n-th convolutional layers to generate a first to an n-th encoded feature maps from the input image; instructing an n-th to a first deconvolutional layers to generate an n-th to a first decoded feature maps from the n-th encoded feature map; if a specific decoded feature map is divided into directions of rows and columns, generating an obstacle segmentation result by referring to a feature of the n-th to the first decoded feature maps; instructing an RPN to generate an ROI bounding box by referring to each anchor box, and losses by referring to the ROI bounding box and its corresponding GT; and backpropagating the losses, to learn the parameters.

Abstract: Automated nuclei area/number estimation utilizes a two-stage estimation framework-area estimation first followed by number estimation. After determining area information, each local patch's shape features are able to be extracted to define a local voting rule. The resulting voting score determines the strength of each local voting peak. The number of voting peaks is exactly the number of nuclei.