Abstract: Systems and processes for identifying a pointer in an image of an analog instrument are provided herein. An instrument contour in the image corresponding to the analog instrument may be identified. A plurality of candidate pointer contours in the image may be identified and screened using one or more geometric property screening techniques including an evaluation of a geometric area, a distance parameter, and/or a gravity center of the plurality of candidate pointer contours. Principal component analysis (PCA) may be performed to select an identified pointer contour from among the reduced plurality of candidate pointer contours. A linear regression model may be applied to pixel points in the contour area of the identified pointer contour and a slope and angle of an associated pointer represented by the identified pointer contour may be determined based on an output of the linear regression model.

Abstract: System and method for joint refinement and perception of images are provided. A learning machine employs an image acquisition device for acquiring a set of training raw images. A processor determines a representation of a raw image, initializes a set of image representation parameters, defines a set of analysis parameters of an image analysis network configured to process the image's representation, and jointly trains the set of representation parameters and the set of analysis parameters to optimize a combined objective function. A module for transforming pixel-values of the raw image to produce a transformed image comprising pixels of variance-stabilized values, a module for successively performing processes of soft camera projection and image projection, and a module for inverse transforming the transformed pixels are disclosed. The image projection performs multi-level spatial convolution, pooling, subsampling, and interpolation.

Abstract: A method configured to implemented on at least one map processing device for determining new roads on a map includes obtaining a first road network image of a region, the first road network image including a first plurality of roads. The method also includes determining a second road network image of the region based on a map of the region, the second road network image including a second plurality of roads that are not present in the first road network image. The method further includes determining a third road network image by concatenating the first road network image and the second road network image. The method still further includes determining a fourth road network image of the region by processing the third road network with at least one convolution layer, the fourth road network including the second plurality of roads.

Abstract: Technologies for analyzing various features detected in a face detected in an image. Such features may include at least the eyes, eyebrows, node, and mouth of a face. Such analyzing may include scoring aspects of these features. Such scores may be weighted. The analysis may be used to determine expressions and/or poses of the face, as well as indicate if the face is smiling, frowning, or neutral. An overall quality score for the face may also be determined and provided based on the aforementioned analysis as well as whether or not the face is near an edge of the image or cut-off in the image. Finally, a face signature may be determined that uniquely identifies the face, at least within the scope of the analyzed features. Such face signatures may be used to detect a similar face in other images.

Abstract: Methods and systems are provided for displaying pertinent data, using augmented reality, to supplement information in images provided from a mobile communication device. The images can be analyzed to find one or more matched objects, and it can be determined whether recognized target components from the matched objects match one or more known patterns. Pertinent data pertaining to the known patterns can be sent to the mobile communication device so that it can be displayed using augmented reality to supplement information in the images.

Abstract: A method and system of stitching a plurality of image views of a scene, including grouping matched points of interest in a plurality of groups, and determining a similarity transformation with smallest rotation angle for each grouping of the matched points. The method further includes generating virtual matching points on non-overlapping area of the plurality of image views and generating virtual matching points on overlapping area for each of the plurality of image views.

Abstract: A system includes a processor and memory storing instructions that when executed by the processor configure the processor to receive an image of a metallographic sample of a product from a microscope, the image including dendritic structures in the metallographic sample; select a portion of the image; and perform morphological operations on the selected portion of the image to transform the image. The instructions configure the processor to overlay a plurality of concentric circles on the transformed image, measure circumferences of the circles, and count the number of intercepts of the circles and boundaries of the dendritic structures in the transformed image. The instructions configure the processor to determine parameters of the dendritic structures based on the measured circumferences and the number of intercepts. The instructions configure the processor to determine structural properties of the product based on the parameters of the dendritic structures.

Abstract: Techniques are provided for using one or more machine learning systems to process input data including image data. The input data including the image data can be obtained, and at least one machine learning system can be applied to at least a portion of the image data to determine at least one color component value for one or more pixels of at least the portion of the image data. Based on application of the at least one machine learning system to at least the portion of the image data, output image data for a frame of output image data can be generated. The output image data includes at least one color component value for one or more pixels of the frame of output image data. Application of the at least one machine learning system causes the output image data to have a reduced dimensionality relative to the input data.

Abstract: Provided is a technique capable of simultaneously satisfying two requests of removing a speckle and clarifying a tissue structure. A noise in an ultrasonic image is removed, and a morphology processing is performed on a noise-removed image. The morphology processing includes a first calculation of performing dilation and erosion and a second calculation of performing opening and closing, and determines a value of a structural element used in the second calculation of the morphology by using a result of the first calculation performed on the noise-removed image.

Abstract: This disclosure relates to method of identifying a gesture from a plurality of gestures using a reservoir based convolutional spiking neural network. A two-dimensional spike streams is received from neuromorphic event camera as an input. The two-dimensional spike streams associated with at least one gestures from a plurality of gestures is preprocessed to obtain plurality of spike frames. The plurality of spike frames is processed by a multi layered convolutional spiking neural network to learn plurality of spatial features from the at least one gesture. A filter block is deactivated from the plurality of filter blocks corresponds to at least one gesture which are not currently being learnt. A spatio-temporal features is obtained by allowing the spike activations from CSNN layer to flow through the reservoir. The spatial feature is classified by classifier from the CSNN layer and the spatio-temporal features from the reservoir to obtain set of prioritized gestures.

Abstract: A method of encoding a watermark into a digital image is provided. The method includes partitioning an image into a plurality of blocks of a same size; accumulating the plurality of blocks of the same size into a single block image; performing a Fourier transformation on the single block image to obtain a two-dimensional Fourier spectrum defined by Fourier coefficients at different positions of a Fourier domain; inserting a watermark into a frequency domain of the two-dimensional Fourier spectrum by modifying the two-dimensional Fourier spectrum as a function of watermarking coefficients in the watermark, to obtain a modified Fourier spectrum; performing an inverse Fourier transformation on the modified Fourier spectrum to obtain a watermarked image; copying the watermarked image horizontally and vertically into a plurality of copied watermarked images; and splicing the plurality of copied watermarked images into a reconstituted watermark image.

Abstract: Provided in a fingerprint image preprocessing method including receiving an input fingerprint image, performing a short-time Fourier transform (STFT) on the input fingerprint image to obtain a transformed fingerprint image, comparing the input fingerprint image and the transformed fingerprint image, and generating a combined image by combining the input fingerprint image and the transformed fingerprint image based on a result of the comparing.

Abstract: A system and techniques that use one or more machine learning models to predict a dense depth map (e.g., of depth values for all pixels or at least more pixels than a sparse estimation source (e.g., SLAM)). In some implementations, the machine learning model includes two sub models (e.g., neural networks). The first machine learning model predicts computer vision data such as semantic labels and surface normal directions from an input image. This computer vision data will be used to add to or otherwise improve sparse depth data. Specifically, a second machine learning model takes the semantic labels and surface normal directions from and sparse depth data (e.g., 3D points) from a sparse point estimation source (e.g., SLAM) as inputs and outputs a depth map. The output depth map effectively densities the initial depth data (e.g., from SLAM) by providing depth data for additional pixels of the image.

Abstract: Example-based affine registration is provided. In various embodiments, a plurality of training images is read. A predetermined affine transform is read for each of the plurality of training images. Each affine transform maps its associated image to a template. Weights are determined for each of the plurality of training images. The weights are determined to minimize a difference between the test image and a weighted linear combination of the training images. An affine transform is determined mapping the test image to the template by computing a weighted linear combination of the affine transforms using the weights.

Abstract: An electronic device is provided. The electronic device includes a touchscreen display, a pressure sensor positioned to sense external pressure against the display, a fingerprint sensor positioned to detect a fingerprint on at least a portion of the display, a processor electrically coupled to the display, the pressure sensor, and the fingerprint sensor, and a memory electrically coupled to the processor, in which the memory stores at least one registered fingerprint. The processor is configured to sense pressure of a user's finger against the display using the pressure sensor, upon sensing of the pressure, activate the fingerprint sensor, detect a fingerprint of the finger using the fingerprint sensor, determine whether the detected fingerprint is matched with any of the at least one registered fingerprint, and perform a preselected function without further requiring authentication, when the detected fingerprint is matched with any of the at least one registered fingerprint.

Abstract: A method for facial image recognition is provided. A plurality of original facial images are received. A plurality of standard facial images corresponding to the original facial images are generated through a standard face generation model. A recognition model is trained by using the original facial images and the standard facial images. The recognition model is tested by using the original facial image test set and a standard facial image test set until the recognition model recognizes that the first accuracy rate of the original facial image test set is higher than a first threshold value and the second accuracy rate of the standard facial image test set is higher than a second threshold value. The original facial image test set is composed of the original facial images obtained by sampling, and the standard facial image test set is composed of the standard facial images obtained by sampling.

Abstract: A spacer supportability evaluation method and device as well as a computer readable storage medium are provided. The method includes acquiring initial distribution images of spacers and corresponding support pads on a substrate, performing binary grayscaling processing to obtain distribution images of spacers and corresponding support pads, obtaining two binary matrices according to the distribution images, subjecting the two binary matrices to convolution in a spatial domain or to multiplication in a frequency domain to obtain an equivalent support matrix, calculating a number of elements in the equivalent support matrix whose values are a first value to obtain a number of supported pixels. The supportability of spacers is evaluated by acquiring parameters or design drawings of the spacers to calculate suitable size and positional arrangement of each spacer, improving the supportability of spacers and keeps the cell gap of the liquid crystal cell stable and uniform.

Abstract: A device includes at least one processor and memory including instructions that when executed by the at least one processor, cause the at least one processor to scan an input image including pixels that have either a first state or a second state opposite the first state, initiate a first cluster and a first polygon that belongs to the first cluster upon encountering a first pixel of the input image that has the first state, execute a first set of operations to form the first polygon, generate an indication of one or more characteristics of the input image based on at least the first cluster, and output the indication.

Abstract: Methods and systems for image analysis are provided, and in particular for identifying a set of base-calling locations in a flow cell for DNA sequencing. These include capturing flow cell images after each sequencing step performed on the flow cell, and identifying candidate cluster centers in at least one of the flow cell images. Intensities are determined for each candidate cluster center in a set of flow cell images. Purities are determined for each candidate cluster center based on the intensities. Each candidate cluster center with a purity greater than the purity of the surrounding candidate cluster centers within a distance threshold is added to a template set of base-calling locations.

Abstract: Systems and methods are provided for processing a set of multiple serially acquired magnetic resonance spectroscopy (MRS) free induction decay (FID) frames from a multi-frame MRS acquisition series from a region of interest (ROI) in a subject, and for providing a post-processed MRS spectrum. Processing parameters are dynamically varied while measuring results to determine the optimal post-processed results. Spectral regions opposite water from chemical regions of interest are evaluated and used in at least one processing operation. Frequency shift error is estimated via spectral correlation between free induction decay (FID) frames and a reference spectrum. Multiple groups of FID frames within the acquired set are identified to different phases corresponding with a phase step cycle of the acquisition. Baseline correction is also performed via rank order filter (ROF) estimate and a polynomial fit.