Abstract: A system for detecting clouds and cloud shadows is described. In one approach, clouds and cloud shadows within a remote sensing image are detected through a three step process. In the first stage a high-precision low-recall classifier is used to identify cloud seed pixels within the image. In the second stage, a low-precision high-recall classifier is used to identify potential cloud pixels within the image. Additionally, in the second stage, the cloud seed pixels are grown into the potential cloud pixels to identify clusters of pixels which have a high likelihood of representing clouds. In the third stage, a geometric technique is used to determine pixels which likely represent shadows cast by the clouds identified in the second stage. The clouds identified in the second stage and the shadows identified in the third stage are then exported as a cloud mask and shadow mask of the remote sensing image.

Abstract: The present invention relates to a biometric authentication apparatus and a biometric authentication method. In the present invention, provided is the authentication apparatus and authentication method of, while placing an object for biometric authentication on a space between a first electrode and a second electrode, applying a square wave with a single period to the first electrode and performing biometric authentication using a measurement voltage width, which is a difference between a maximum value (a maximum measurement voltage value) and a minimum value (a minimum measurement voltage value) of voltage measured at the first electrode, and an arriving time required for the voltage measured at the first electrode to reach a particular range of the maximum measurement voltage value from the minimum measurement voltage value. The biometric authentication apparatus and method according to the present invention possibly provide a human body detect function for personal authentication with high reliability.

Abstract: The present disclosure relates to an image processing device and method which are capable of suppressing an increase in a storage capacity necessary for encoding and decoding. A motion compensating unit that performs motion compensation in decoding of a current layer and a first compressing unit that compresses a motion vector of the current layer that is reconstructed by the motion compensating unit and used for the motion compensation in decoding of another layer are provided. Alternatively, a motion predicting/compensating unit that performs motion prediction and compensation in encoding of a current layer and a first compressing unit that compresses a motion vector of the current layer that is generated by the motion predicting/compensating unit and used in the motion prediction and compensation in encoding of another layer are provided. For example, the present disclosure can be applied to an image processing device.

Abstract: A dependency indication is signaled within the beginning of a packet, that is, within the adjacent of a slice header to be parsed or a parameter set. This is achieved, for example, by including the dependency indication at the beginning of the slice header, preferably after a syntax element identifying the parameter set and before the slice address, by including the dependency indication before the slice address, by providing the dependency indication to a NALU header using a separate message, or by using a special NALU type for NALUs carrying dependent slices.

Abstract: The present disclosure provides an image encoding method and apparatus and an image decoding method and apparatus for generating a reconstructed image having a minimized error between an original image and the reconstructed image at a high bit depth and a high bit rate. The image decoding method includes parsing offset values and an offset type from a bitstream; parsing a class based on the offset type from the bitstream; selecting a plurality of pixels from among pixels adjacent to a reconstructed pixel, based on the class, and calculating a reference value based on sample values of the plurality of pixels; determining a category based on a magnitude of a difference value between the reference value and a sample value of the reconstructed pixel; selecting an offset value from among the offset values, the offset value corresponding to the category; and compensating for the sample value of the reconstructed pixel by using the selected offset value.

Abstract: An image encoding apparatus based on a system on chip (SoC) which encodes a residual block of a current block including a first value calculator configured to calculate a first value for the residual block in a space domain, a comparator configured to decide whether to transform the residual block into a transform domain according to a size of the first value, and a transform unit configured to transform the residual block into the transform domain according to a decision on whether to transform.

Abstract: Disclosed systems and methods categorize text regions of an electronic document into document object types based on a combination of semantic information and appearance information from the electronic document. A page segmentation application executing on a computing device accesses textual feature representations that represent text portions in a vector space, where a set of pixels from the page is mapped to a textual feature representation. The page segmentation application generates a visual feature representation, which corresponds to an appearance of a document portion including the set of pixels, by applying a neural network to the page of the electronic document. The page segmentation application generates an output page segmentation of the electronic document by applying the neural network to the textual feature representation and the visual feature representation.

Abstract: The present invention provides an apparatus and a method for enhancing spatial resolution of a CT image and a CT imaging system, the method comprising: acquiring an original CT projection curve; performing deconvolution for projection data on the original CT projection curve in a tube sampling direction or a texture direction of the original CT projection curve; and reconstructing an image according to the projection data after deconvolution.

Abstract: The present disclosure relates to a font recognition system that employs a multi-task learning framework and adversarial training to improve font classification and remove negative side effects caused by intra-class variances of glyph content. For example, in one or more embodiments, the font recognition system adversarial trains a font recognition neural network by minimizing font classification loss while at the same time maximizing glyph classification loss. By employing an adversarially trained font classification neural network, the font recognition system can improve overall font recognition by removing the negative side effects from diverse glyph content.

Abstract: The present disclosure discloses a photo processing method and an apparatus for grouping photos into photo albums based on facial recognition results. The method includes: performing face detection on multiple photos, to obtain a face image feature set, each face image feature in the face image feature set corresponding to one of the multiple photos; determining a face-level similarity for each pair of face image features in the face image feature set; determining a photo-level similarity between each pair of photos in the multiple photos in accordance with their associated face-level similarities; generating a photo set for each target photo in the multiple photos, wherein any photo-level similarity between the target photo and another photo in the photo set exceeds a predefined photo-level threshold; and generating a label for each photo set using photographing location and photographing time information associated with the photos in the photo set.

Abstract: The disclosure discloses a method and device for image recognition. The method comprises: placing instruction blocks in a required order; acquiring an encoded instruction block image that represents the instruction block pattern and the placement order; recognizing a single instruction block image from the encoded instruction block image; comparing the single instruction block image with a standard instruction block image to obtain the matching degree of the instruction block; determining, according to the matching degree, standard instruction blocks corresponding to respective instruction block images; sorting the instruction block images according to the coordinates in the encoded instruction block image; parsing the instruction block images according to the sorting order to obtain instructions. The method for image recognition can help children learning concepts such as manipulation, use and programming of intelligent programmable devices more easily.

Abstract: A method of quality inspection is performed by a robotic arm that includes a plurality of segments, a camera at an end of the robotic arm, and a plurality of joints connecting two segments of the plurality of segments. The method includes (i) inspecting, via the camera, a surface of a product with the camera positioned at a first position, (ii) based on the inspecting, identifying: (a) an area of interest on the surface of the product, and (b) a relative location of the area of interest on the surface, (iii) positioning, based on the relative location of the area of interest on the surface, the camera at a second position, and (iv) inspecting, via the camera, the area of interest on the surface of the product with the camera positioned at the second position. Inspecting the area of interest includes inspecting a subset of the surface of the product.

Abstract: A method for detecting a design-impacting defect in an integrated circuit substrate is disclosed. In one implementation, a controller determines a distribution of intended geometric features in a design window of the integrated circuit substrate based on proximities of a plurality of points of interest in the design window to the intended geometric features. The controller obtains a set of intended contours from the distribution. The controller obtains a set of imaged contours from one or more images of the integrated circuit substrate. The controller compares the set of imaged contours to the set of intended contours to obtain a set of potential design-impacting defects in the intended geometric features. The controller determines a probability that a potential design-impacting defect from the set of potential design-impacting defects is a valid design-impacting defect. The controller takes a corrective action based on the determined probability.

Abstract: A system and method that gathers and labels images for use by a machine learning algorithm for image classification is disclosed. The method includes acquiring a preview image including a portion of a scene, generating a user interface including the preview image, performing object detection on the preview image to detect a set of items, adding, to the first preview image, a set of regions of interest based on the object detection, each region of interest highlighting a location of a corresponding item in the set of items, receiving an image corresponding to the preview image, determining an object identifier associated with an item, and labeling the item in the image using the object identifier and a region of interest corresponding to the item.

Abstract: As set forth herein low energy signal data and high energy signal data can be acquired. A first material decomposed (MD) image of a first material basis and a second material decomposed (MD) image of a second material basis can be obtained using the low energy signal data and the high energy signal data. At least one of the first or second MD image can be input into a guide filter for output of at least one noise reduced and cross-contamination reduced image. A computed tomography (CT) imaging system can be provided that includes an X-ray source and a detector having a plurality of detector elements that detect X-ray beams emitted from the X-ray source. Low energy signal data and high energy signal data can be acquired using the detector.

Abstract: According to an aspect of an embodiment, a method may include receiving, at a first interface element of a user interface, a first user input regarding a degree of stereoscopic depth rendered in a stereoscopic image. The method may also include adjusting the stereoscopic depth based on the first user input. Additionally, the method may include receiving, at a second interface element of the user interface, a second user input regarding adjustment of a z-plane position of the stereoscopic image and adjusting the z-plane position based on the second user input. The method may further include generating the stereoscopic image based on the adjustment of the stereoscopic depth and the adjustment of the z-plan position.

Abstract: Using an image capturing device, multiple images can be obtained at different angles with respect to a surface of a target coating. In one computer-implemented embodiment, a method involves using a filtering technique to perform an image analysis on the obtained images to determine the presence of sparkle points within the images. A color attribute analysis can then be performed to determine various color attributes associated with the determined sparkle point. A sparkle color distribution can then be calculated in response to the color attribute analysis. A coating formulation can then be generated, in association with the calculated sparkle color distribution, which is the same or substantially similar to the target coating.

Abstract: A method, computer readable medium, and system are disclosed for training a neural network. The method includes the steps of selecting an input sample from a set of training data that includes input samples and noisy target samples, where the input samples and the noisy target samples each correspond to a latent, clean target sample. The input sample is processed by a neural network model to produce an output and a noisy target sample is selected from the set of training data, where the noisy target samples have a distribution relative to the latent, clean target sample. The method also includes adjusting parameter values of the neural network model to reduce differences between the output and the noisy target sample.

Abstract: Presented herein are efficient and reliable systems and methods for calculating and extracting three-dimensional central axes of bones of animal subjects—for example, animal subjects scanned by in vivo or ex vivo microCT platforms—to capture both the general and localized tangential directions of the bone, along with its shape, form, curvature, and orientation. With bone detection and segmentation algorithms, the skeletal bones of animal subjects scanned by CT or microCT scanners can be detected, segmented, and visualized. Three dimensional central axes determined using these methods provide important information about the skeletal bones.

Abstract: Provided is a re-encoding method including obtaining a first quantization table from a bitstream including an image encoded using the first quantization table; obtaining a second quantization table based on a pattern representing a size distribution of values of elements of the first quantization table, the second quantization table including elements respectively corresponding to the elements of the first quantization table; and re-encoding a reconstructed image by using the second quantization table, the reconstructed image being obtained by decoding the encoded image by using the first quantization table.