Abstract: Disclosed is a technique capable of efficiently storing and retaining characteristic data (a restoration filter or the like) of an optical system used for a restoration process in a storage unit with limited storage capacity in consideration of the degree of image restoration. An image processing device includes a characteristic data storage unit 42 which is capable of storing characteristic data of a plurality of types of optical systems, and a restoration processing unit which subjects source image data to a restoration process using a restoration filter based on a point spread function of an optical system to acquire recovered image data. In case where storing new characteristic data in the characteristic data storage unit, characteristic data which is stored in the characteristic data storage unit 42 is controlled based on a restoration evaluation value which is allocated to characteristic data according to the type of optical system.

Abstract: A handheld imaging device has a data receiver that is configured to receive reference encoded image data. The data includes reference code values, which are encoded by an external coding system. The reference code values represent reference gray levels, which are being selected using a reference grayscale display function that is based on perceptual non-linearity of human vision adapted at different light levels to spatial frequencies. The imaging device also has a data converter that is configured to access a code mapping between the reference code values and device-specific code values of the imaging device. The device-specific code values are configured to produce gray levels that are specific to the imaging device. Based on the code mapping, the data converter is configured to transcode the reference encoded image data into device-specific image data, which is encoded with the device-specific code values.

Abstract: A first set of attributes (e.g., style) is generated through pre-trained single column neural networks and leveraged to regularize the training process of a regularized double-column convolutional neural network (RDCNN). Parameters of the first column (e.g., style) of the RDCNN are fixed during RDCNN training. Parameters of the second column (e.g., aesthetics) are fine-tuned while training the RDCNN and the learning process is supervised by the label identified by the second column (e.g., aesthetics). Thus, features of the images may be leveraged to boost classification accuracy of other features by learning a RDCNN.

Abstract: A fingerprint sensing system comprising a device connection interface including a device reference potential input, a sensing arrangement, and sensing reference potential providing circuitry. The sensing arrangement includes multiple sensing structures and read-out circuitry connected to each of the sensing structures.

Abstract: Methods and systems for capturing motion and/or determining the shapes and positions of one or more objects in 3D space utilize cross-sections thereof. In various embodiments, images of the cross-sections are captured using a camera based on reflections therefrom or shadows cast thereby.

Abstract: The present disclosure relates to digital watermarking. One claim recites a method to detect two or more different digital watermarks in media. The method includes: receiving captured imagery of the media, the captured imagery comprising a plurality of image frames; for a first image frame applying a first watermark detector to search for a first digital watermark hidden within the first image frame, in which an electronic processor is programmed as the first watermark detector; and for a second image frame applying a second, different watermark detector to search for a second, different watermark hidden within the second image frame, in which an electronic processor is programmed as the second watermark detector. Other claims and combinations are provided too.

Abstract: Local models learned from anomaly detection are used to rank detected anomalies. The local model patterns are defined from image feature values extracted from an image field of video image data with respect to different predefined spatial and temporal local units, wherein anomaly results are determined by fitting extracted image features to the local model patterns. Image features values extracted from the image field local units associated with anomaly results are normalized, and image feature values extracted from the image field local units are clustered. Weights for anomaly results are learned as a function of the relations of the normalized extracted image feature values to the clustered image feature values. The normalized values are multiplied by the learned weights to generate ranking values to rank the anomalies.

Abstract: A light guide mechanism of a fingerprint recognition plate includes a frame, a light guide plate, first electrodes, second electrodes and light emitting diodes. The frame has a hollow portion and receiving troughs. Each of the receiving troughs forms a first through hole and a second through hole penetrating the frame. The light guide plate covers the hollow portion. The first electrodes are in the first through holes, and the second electrodes are in the second through holes. Each of the light emitting diodes is in the receiving trough and includes a first electrode pin and a second electrode pin. The first electrode pin pierces the first through hole to be connected to the first electrode. The second electrode pin pierces through the second through hole to be connected to the second electrode. The first electrode pin and second electrode pin are exposed from the bottom of the frame.

Abstract: A system, method, and computer program product for estimating human body pose are described. According to one aspect, a human figure silhouette is segmented from a depth image of a human actor. Contour points are sampled along the human figure silhouette. Inner Distance Shape Context (IDSC) descriptors of the sample contour points are determined and compared to IDSC descriptors of the feature points in an IDSC gallery for similarity. For each of the feature points, the sample contour point with the IDSC descriptor that is most similar to an IDSC of the feature point is identified as that feature point in the depth image. An estimated pose of a human model is estimated based on the detected feature points and kinematic constraints of the human model.

Abstract: By either subsampling neighbor pixels and deriving an illumination variation parameter, or deriving one normalization shift value of two normalization shift values for normalizing parameters which are used when deriving the illumination variation parameter, with a dependency on the other normalization shift value, an amount of calculation for illumination compensation is reduced.

Abstract: A method is provided for generating and revising map geometry based on a received image and probe data. A method may include: receiving probe data from a first period of time, where the probe data from a first period of time is from a plurality of probes within a predefined geographic region; generating a first image of the predefined geographic region based on the probe data from the first period of time; receiving probe data from a second period of time different from the first period of time, where the probe data from the second period of time is from a plurality of probes within the predefined geographic region; generating a second image based on the probe data from the second period of time; comparing the first image to the second image; and generating a revised route geometry based on changes detected between the first image and the second image.

Abstract: A biometric sensor device for measuring structures and properties of an object of organic tissue, especially a fingerprint sensor, includes a plurality of linear sensing arrays adapted to generate a respective plurality of measurement data at a predetermined sampling rate for a predetermined time period, based upon a sliding movement of the organic tissue over the plurality of linear sensing arrays, and a processor adapted to generate a partial image of the organic tissue from each of the respective plurality of measurement data. The processor is further adapted to generate a complete image of the organic tissue by combining the partial image generated from each of the plurality of measurement data, wherein the complete image is larger than each of the partial images, and the predetermined time period is defined so as to measure a section of said organic tissue given by the distance between two linear arrays.

Abstract: Disclosed are a method and apparatus for embedding a graphic image representation into a two dimensional matrix code by modifying the characteristic values of individual pixels in the image according the values of a provided two dimensional matrix code image. The modified character pixel values are determined using an optimization procedure that minimizes a visual distortion with respect to the original graphic image representation while maintaining the value of a probability of error model below a specified limit.

Abstract: Methods, apparatus, systems, and computer-readable media are provided for using a camera of a robot to capture an image of the robot's environment, detecting edges in the image, and localizing the robot based on comparing the detected edges in the image to edges derived from a three-dimensional (“3D”) model of the robot's environment from the point of view of an estimated pose of the robot in the environment. In some implementations, the edges are derived based on rendering, from the 3D model of the environment, a model image of the environment from the point of view of the estimated pose—and applying an edge detector to the rendered model image to detect model image edges from the model image.

Abstract: A mechanism is described for facilitating three-dimensional (3D) depth imaging systems, and morphological and geometric filters for edge enhancement in depth images at computing devices according to one embodiment. A method of embodiments, as described herein, includes detecting an input digital image of an object, the digital image comprising data pixels contaminated by noise and confidence values corresponding to the data pixels, and computing a morphological filter by matching the confidence pixels in the input digital image with a set of matching templates, and using a set of masking templates to determine the data pixels and confidence pixels in the filtered image. The method further include computing an edge filter by performing computation of distances between the data pixels along a plurality of directions to determine an edge direction, and determining the data pixels and and the confidence pixels in a filtered image based on the edge direction.

Abstract: A defog system and a defog method are provided. The defog system includes an illuminance sensor configured to detect ambient illuminance, and at least one processor to implement a level determiner configured to determine a defog level of an input image based on the ambient illuminance. The at least one processor further implements a defogger configured to determine whether to defog the input image based on skewness and kurtosis of a histogram of the input image, and, in response to the defogger determining to defog the input image, determine a fog condition of the input image based on a cumulative probability of the histogram, and defog the input image based on the defog level.

Abstract: A method of tracking objects of a scene is disclosed. The method determines two or more tracks which have merged. Each track is associated with at least one of the objects and having a corresponding graph structure. Each graph structure comprising at least one node representing the corresponding track. A new node representing the merged tracks is created. The graph structures are added as children nodes of the new node to create a merged graph structure. A split between the objects associated with one of the tracks represented by the nodes of the merged graph structure is determined. Similarity between one or more of the nodes in the merged graph structure and foreground areas corresponding to split objects is determined. One of the nodes in the merged graph structure is selected based on the determined similarity. A new graph structure for tracking the objects is created, the new graph structure having the selected node at the root of the new graph structure.

Abstract: A method of extracting a bed area of a bed performed by a computer, includes detecting linear edges which remain stationary among plural images; and selecting a bed area candidate based on lengths of the linear edges and angles between the linear edges which cross to each other in a U shape formed as combinations of the detected linear edges.

Abstract: A method of deblocking an input signal is disclosed. The method generally includes the steps of (A) calculating a plurality of transform coefficients corresponding to each of a plurality of blocks in the input signal at baseband, (B) calculating a plurality of quantization parameters based on the transform coefficients, at least one of the quantization parameters corresponding to each respective one of the blocks and (C) generating an output signal by deblocking the input signal based on the quantization parameters.

Abstract: A system for evaluating a classifier of an image signal processor (ISP) includes (i) a microprocessor and (ii) memory storing training images, the microprocessor being capable of sending each training image to the ISP. The system includes machine-readable instructions stored within the memory and executed by the microprocessor capable of: (i) selecting a subset of images based upon a divider position, (ii) controlling the ISP to classify each image as belonging or not belonging to an object class, (iii) determining a positive-match count, (iv) determining an error count based upon the positive-match count and total number of training images belonging to the object class, (v) repeating, for other divider positions, steps of selecting, controlling, and determining to identify an optimal divider position and a minimum-error count; and (vi) determining the classifier's optimality by comparing the optimal divider position to a predetermined optimal divider position and a predetermined minimum-error count.