71 results about "Reference Region" patented technology

According to embodiments of this disclosure, a method and apparatus for data processing are provided. The method includes: determining a reference region corresponding to a target traffic light indicated by map data in an image coordinate system corresponding to a perceived image; determining a set of candidate objects from a plurality of candidate objects identified based on the perceived image, based on the positional relationship between the reference region and a plurality of candidate regions, wherein the plurality of candidate regions correspond to the plurality of candidate objects identified as traffic lights; determining a first correlation between a first attribute of the target traffic light and a second attribute of the set of candidate objects, wherein the first attribute is determined based on the map data and the second attribute is determined based on the perceived image; and matching the target traffic light with the target objects in the set of candidate objects based on the first correlation. Based on this approach, embodiments of this disclosure can more accurately identify and match traffic lights.

A video decoding method according to an embodiment comprises: obtaining flag information indicating whether MIP is applied and prediction mode information of a current block; when the flag information indicates that MIP is applied and a prediction mode of the current block is an MIP prediction mode, setting a reference region used for MIP on the basis of condition information of the current block; determining a matrix used for MIP; and obtaining a prediction sample of the current block on the basis of the reference region and the matrix.

This invention belongs to the field of medical image processing and deep learning inference technology, specifically relating to a method, device, and storage medium for denoising and enhancing low-dose CT images. The denoising and enhancement method includes the following steps: S1, acquiring a low-dose CT image; S2, calculating the noise level parameter N; S2.1, generating a uniform reference region mask; S2.2, robust noise estimation within the mask; S2.3, robust fusion of the noise level parameter; S3, generating a denoised image I_dn based on the CT image to be processed using a deep learning denoising network; S4, calculating the denoising intensity coefficient s according to the noise level parameter N, and performing weighted fusion of the CT image to be processed and the denoised image I_dn to obtain the output enhanced image. This invention has the advantages of strong adaptability, no need for manual ROI selection, good robustness, and easy deployment.

The present application discloses an encoding method, a decoding method, a code stream, an encoder, a decoder, and a storage medium. The method comprises: determining one or more vector parameters of a current block; determining at least one second reference region on the basis of a first reference region indicated by the one or more vector parameters, wherein the second reference region is obtained by merging first reference regions indicated by at least some vector parameters among the one or more vector parameters; determining one or more prediction modes of the current block on the basis of the at least one second reference region, wherein the one or more prediction modes are determined on the basis of gradient information derived from the at least one second reference region; and determining a prediction sample of the current block on the basis of the one or more prediction modes.

The present invention relates to a camera mirror system, and more particularly, to a camera mirror system that matches screen information such as a screen menu type adjustment method to an image using calibration information for calibrating installation errors of a camera mirror equipped in a vehicle. The camera mirror system includes a camera disposed at a side of the vehicle to photograph a rear or a side rear of the vehicle, a display device that outputs an image photographed by the camera, and a control device that relays the image between the camera and the display device, wherein the control device includes a calibration information calculation section that calculates calibration information for calibrating installation errors of the camera, a region management section that sets an extension region that extends a predetermined reference region in an original image generated by the camera using the calibration information, an image processing section that extracts an extension image corresponding to the extension region in the original image, and a transmission section that transmits a display image corresponding to a display region selected by a user in the extension image to the display device.

This invention discloses a method, apparatus, and storage medium for detecting leftover objects in elevators. The method includes: acquiring raw image data that meets background conditions inside the elevator car; drawing a detection area located at the bottom of the car and a light reference area located on the side wall of the car in the raw image data; calculating a first brightness value of the detection area and a second brightness value of the light reference area in the raw image data; querying one or more background image groups; selecting background image data that matches the brightness of the raw image data from the background image groups based on the first and second brightness values ​​of the raw image data, as target image data; and comparing the detection area of ​​the target image data with the detection area of ​​the raw image data to identify whether there are leftover objects in the car. This embodiment effectively improves the efficiency of detecting leftover objects in elevators.

Provided in the present invention are an automatic color matching configuration method and apparatus suitable for marble. The method comprises: acquiring a to‑be‑repaired-marble image of marble to be repaired, and performing identification to determine a to‑be‑repaired contour in the to‑be‑repaired-marble image; extending outward by a preset distance from the to‑be‑repaired contour, so as to generate a reference contour, wherein the region between the reference contour and the to‑be‑repaired contour is a reference region; using a base coating material to fill a region to be repaired within the to‑be‑repaired contour, so as to form an initial base coating layer; on the basis of the color proportion condition of the reference region, determining a base pixel value, and on the basis of the base pixel value, performing base color matching on the initial base coating layer to obtain a current base coating layer; on the basis of the pattern distribution condition of the reference region, performing regional color matching on the current base coating layer to obtain a repaired region, and determining the marble to be repaired as marble that has been repaired; and acquiring a repaired-marble image of the marble that has been repaired, and on the basis of the repaired-marble image and the to‑be‑repaired-marble image, generating a historical record entry. By means of the present invention, at least the repair effect of marble is improved.

Methods and systems for expanding the drawing surface size across the same or different bitstream layers of a video coding bitstream are described. [Solution] The offset parameter of the fitting window, the reference region of interest (ROI) of the reference layer, and the current ROI of the current layer are received. The width and height of the current ROI and reference ROI are calculated based on the offset parameter, and these are used to generate width and height scaling factors that should be used by the reference picture resampling unit to generate an output picture based on the current ROI and reference ROI.

The application discloses a non-measured area electromagnetic clutter deduction method and system based on similar terrain areas, and the method comprises the following steps: acquiring geographic information data of a target non-measured area, and matching one or more reference areas with similar terrain characteristics; extracting measured electromagnetic clutter data and corresponding environmental parameters of each reference area, and constructing a terrain-electromagnetic correlation feature mapping model by using a deep feature learning network; dynamically generating a correction factor suitable for the area according to the terrain-electromagnetic correlation feature mapping model; based on the correction factor, deducing the spatial clutter intensity distribution of the target non-measured area under a specified frequency band and polarization mode; performing statistical feature conformity test and spatial continuity optimization on the spatial clutter intensity distribution, and finally outputting the electromagnetic clutter deduction result of the non-measured area. By using the embodiment of the application, the accuracy and spatial continuity of the non-measured area electromagnetic clutter deduction can be improved, and the dependence on field measurement can be reduced.

The application discloses a template-based adaptive cross-component prediction method, which comprises the following steps: judging whether the reconstructed pixels of the neighboring area of a central unit are available, and obtaining a first judgment result; selecting a target prediction mode and obtaining a target reference area; combining the target reference area with the first judgment result to obtain a target extension area corresponding to the target reference area; obtaining a cross-component prediction model according to the target reference area and the target extension area; establishing a target template corresponding to the target prediction mode; obtaining a SATD cost according to the target template and the cross-component prediction model, and obtaining a first prediction model; obtaining a target RDO cost corresponding to the first prediction model, comparing the target RDO cost with a current optimal RDO cost, and obtaining a second prediction model as an optimal chroma prediction model. The application can obtain the optimal prediction model of the central unit, improves the coding performance, and can be widely applied in the technical field of video processing.

The application discloses a vehicle-mounted camera anomaly detection method and device, equipment and a storage medium, relates to the field of vehicle engineering, and is applied to a target cloud and comprises the following steps: sending an anomaly detection instruction to a target vehicle terminal of a target vehicle based on a preset trigger condition to obtain a target media file; analyzing a target statistical characteristic of the target media file to obtain a first analysis result, determining a target detection model based on a target camera corresponding to the target media file, and performing reference area checking on the target media file by using the target detection model to obtain a corresponding second analysis result; determining a first confidence degree corresponding to the first analysis result and a second confidence degree corresponding to the second analysis result by using a target confidence degree algorithm, and fusing the first analysis result and the second analysis result based on the first confidence degree and the second confidence degree to obtain an anomaly detection result corresponding to the target camera. The application improves the accuracy and efficiency of vehicle-mounted camera anomaly detection.

The technology disclosed in the present description relates to a technology for determining a position of a substrate with high accuracy. A position determination method includes: a step of setting a region including an end portion of a substrate in a reference image as a reference region and detecting a pixel position of the end portion of the substrate in the reference region as a reference pixel position; a step of setting a region including the end portion of the substrate in a comparison image as a comparison region and detecting a pixel position of the end portion of the substrate in the comparison region as a comparison pixel position; and a step of determining whether or not a difference between the reference pixel position and the comparison pixel position exceeds a predetermined threshold.

A method for manufacturing a connecting part between a wing box and a leading edge module. The method comprises performing a three-dimensional scanning of a first representative part of the wing box to obtain a first computer model and performing a three-dimensional scan of a second representative part of the leading edge module in order to obtain a second computer model. The first part comprises a first reference region and a first connecting region, the second part comprises a second reference region and a second connecting region. A relative position of the first reference region is defined with respect to the reference region. From the first and second computer model, the connecting part is configured to connect the first and the second connecting region such that the first and the second reference region are located in said relative position. The connecting part is then manufactured.

A video decoding method according to the present disclosure involves: determining, on the basis of prediction mode information, whether a prediction mode of the current block corresponds to an MIP mode; when the prediction mode of the current block corresponds to the MIP mode, acquiring prediction samples of the current block by determining a matrix and a reference region used for MIP on the basis of the size of the current block; and when one or more intra prediction mode indexes among a plurality of intra prediction mode indexes are set to indicate the MIP mode, determining the prediction mode of the current block as the MIP mode if intra prediction mode information about the current block indicates the MIP mode.

The present embodiment provides a method and device for video coding using intra template matching prediction. In the present embodiment, an image decoding device constructs block vector candidates of a current block on the basis of a template cost indicating a difference between a current template of the current block and a reference template within a predefined reference region. Here, the block vector candidate indicates a reference block corresponding to the reference template, the current template includes pre-reconstructed neighboring pixels of the current block, and the reference template includes neighboring pixels of the reference block. The image decoding device constructs a candidate list for applying a fusion mode, on the basis of the block vector candidates or template costs of the block vector candidates. Here, the candidate list includes new block vector candidates for applying the fusion mode. The image decoding device calculates weights for the new block vector candidates, and predicts the current block on the basis of the weights and the new block vector candidates.

This electronic device may comprise a first volatile memory and at least one processor including a compression circuit, wherein the compression circuit includes a second volatile memory, a first processing circuit and a second processing circuit, the first processing circuit comparing, in order to compress a first segment, the bit sequence of the first segment, which has a designated size and is included in a reference region, with the bit sequence of the first segment positioned next to the reference region. The reference region can be changed within the second volatile memory on the basis of the comparison. The first processing circuit can release, from the second volatile memory, a portion of the first segment in which the reference region has been changed on the basis of the comparison. The second processing circuit can store a portion of the second segment of information in a second volatile memory, which is a portion in which the first segment of the information was stored. Prior to completion of the comparison of the first segment of the information, the second processing circuit can use the reference region to begin comparison of a portion of the second segment of information stored in the second volatile memory in order to compress the second segment of information.

This disclosure provides a quantum chip wiring method, apparatus, and training method for a pairing neural network model. The wiring method includes mapping the coordinate information of components in the quantum chip from their positions to a preset reference region; grouping the starting interfaces of qubits within the reference region according to predetermined rules; grouping and sorting the terminal interfaces of the quantum chip; assigning a corresponding terminal interface group to each starting interface group based on a preset neural network model, and determining the pairing relationship between the starting interfaces and terminal interfaces within each group to output the pairing results; and performing serial wiring based on a preset algorithm to obtain the wiring path. This disclosure decouples pin pairing from path planning, automatically generates pin connection relationships using a deep learning model, simplifies path search to point-to-point wiring, significantly improves the automation and efficiency of quantum chip wiring, and can adapt to quantum chips of different sizes and layouts.

The present invention provides an image processing device, an endoscope system, an image processing method, and a program that enable oxygen saturation imaging with correction for in-plane unevenness caused by imaging system factors in endoscopes. [Solution] The image processing device according to this disclosure acquires a first spectral image obtained by imaging a living organism irradiated with a first light using an endoscope, acquires a second spectral image obtained by imaging a living organism irradiated with a second light having a different wavelength band from the first light using an endoscope, corrects the brightness distribution of the first spectral image with first correction data stored in advance in association with the endoscope, outputs an image showing the distribution of oxygen saturation of blood in the living organism from the corrected first spectral image and the second spectral image, and calculates the first correction data using a first reference spectral image obtained by imaging a reference region with a known brightness distribution using an endoscope and the brightness distribution of the reference region.

Decoding using extensive warp prediction filters includes obtaining reconstructed block data based on a warp motion vector for a center pixel of a predicted block in an available reference area from a reconstructed reference frame. Generating an intermediate predicted block by, on a per-pixel basis, obtaining a horizontal offset value, identifying a horizontal warp filter in accordance with the horizontal offset value, and applying the horizontal warp filter to the reference area to obtain an intermediate predicted pixel. Obtaining a predicted block by, on a per-pixel basis, obtaining a vertical offset value, identifying a vertical warp filter in accordance with the vertical offset value, and applying the vertical warp filter to the intermediate predicted block to obtain a predicted pixel.

The present application relates to the field of encoding and decoding technology and provides a template matching-based prediction method, apparatus, device, and storage medium. The template matching-based prediction method comprises the steps of: determining a first initial search point based on a first vector included in a first vector prediction (VP) within a VP list of the current block; searching for a second vector of the current block in a reference area of ​​the current block using a template formed of reconstructed pixels around the current block based on the first initial search point; and determining a prediction block of the current block based on the second vector. Both the encoder side and the decoder side may determine the prediction block of the current block using the above method. In other words, since the encoder side may not transmit motion information to the decoder side, the bit rate occupied during the transmission of motion information can be saved, and accordingly, the bit rate can be improved.

Provided are an encoding method, a decoding method, an encoder, a decoder, a storage medium, and a bitstream. The decoding method comprises: determining a first block vector of a current block; determining a first reference region of the current block on the basis of the first block vector; determining a predicted value of the current block on the basis of the first reference region; and determining a reconstructed value of the current block on the basis of the predicted value of the current block.

A robot system (1, 2, 3, 4), according to one embodiment, may comprise: a main body (11, 11'); an end effector (122) connected to the main body (11, 11') by means of a plurality of joints (121); a calibration tool (123) installed on the end effector (122) and having at least a portion thereof formed in a spherical shape; and a calibration jig (14, 14a, 14b, 14', 24, 34, 44) disposed in a position in which the calibration tool (123) may be in contact, and comprising a plurality of reference areas (a1, a2, a3, a4) in which the calibration tool (123) may be in three-point contact.

A block vector predictor (BVP) may be adjusted to provide a more accurate prediction of a block vector (BV). A coder (e.g., encoder or decoder) may determine that the at least one component of the BVD may be equal to zero (or, e.g., null), for example, based on a position of the Current Block (CB) relative to a boundary of the reference region. Consequently, the encoder may skip signaling, and the decoder may skip parsing, of the at least one component of the BVD in the bitstream.

This invention provides a fabric performance testing device based on intelligent sensors, relating to the field of material testing technology. It includes an intelligent sensing acquisition module for collecting reflection, scattering, and transmission signals generated by the detection area and the reference area under the action of detection light, and generating corresponding raw test data. This intelligent sensor-based fabric performance testing device establishes a detection benchmark by setting a detection area and a reference area within the same section of the fabric to be tested, and separately collecting their reflection, scattering, and transmission signals under preset testing conditions. The raw test data of the detection area is then corrected, thereby achieving stable acquisition of relevant performance parameters of the fabric's physical properties. This reduces the impact of light source fluctuations, environmental changes, and fabric surface differences on the test results, improving the accuracy, repeatability, and comparability of the test results.

This invention relates to the technical field of testicular ultrasound image denoising, specifically to an automatic optimization method and system for testicular ultrasound images. The method involves acquiring testicular ultrasound images and dividing them into regions. A reference region is selected for Canny edge detection, and multiple sets of edge pixels are obtained by iteratively adjusting a high threshold. An edge variation factor is calculated based on the total number of pixels in the region, the number of reference edge points, and the number of newly added points in the neighborhood during iteration. By analyzing the gradient direction differences and positional distribution of each edge point, pixels with approximate gradient directions are determined. A texture descriptor is obtained by combining the direction differences, the number of approximate points, and the neighborhood gradient differences. The diffusion coefficient of each edge point is calculated by integrating gradient features, texture descriptors, and the edge variation factor, ultimately achieving image denoising. This invention can select an appropriate diffusion coefficient to remove noise from testicular ultrasound images while improving image quality and detail.

A method for improving Local Illumination Compensating applied to a video decoder is provided. The method includes: enabling a Position-Dependent Local Illumination Compensating (PDLIC) for a current coding block; determining a template for the current coding block; determining a reference block related to the current coding block, and reference areas for the reference block, based on the current coding block and the template, wherein coordination of reference samples of the reference block map coordination of samples of the current coding block; determining Local Illumination Compensation parameters based on samples of the template, reference samples of the reference area, and position information of samples in the template; performing the Position-Dependent Local Illumination Compensating for the current coding block based on the Local Illumination Compensation parameters.