144 results about "Hybrid image" patented technology

Various structures for a hybrid pixel array are disclosed. The hybrid pixel array includes a combination of square photodiode (PD) pixel structures and split PD pixel structures. Square PD pixel structures include one photodiode per pixel unit while split PD pixel structures include two photodiodes per pixel unit. In certain instances, square PD pixel structures are used for green light pixels and split PD pixel structures are used for blue and red light pixels in the hybrid pixel array. The combination of square PD pixel structures and split PD pixel structures provides autofocus capability with higher signal strengths. Various techniques for row addressing and readout from the hybrid pixel array are also disclosed.

This invention provides an image processor, a processing method, a storage medium, and a mixed reality display system. The image processor includes: a first color management module, whose input is connected to a rendering module, for acquiring virtual layer data from the rendering module and performing first color processing on it; a second color management module, whose input is connected to a camera module, for acquiring real-world layer data from the camera module and performing second color processing on it; a layer blending module, whose inputs are respectively connected to the first and second color management modules, for performing layer blending processing on the color-processed virtual layer data and real-world layer data to determine mixed image data; and a third color management module, whose input is connected to the layer blending module, for acquiring mixed image data from the layer blending module, performing third color processing on it, and outputting the third-color-processed mixed image data to a mixed reality display.

The present disclosure provides a network model training method, a data processing method, and an apparatus. The network model training method comprises: acquiring target sample data, wherein the target sample data comprises text sample data and image sample data; inputting the target sample data into a network model to be trained to obtain a sample recognition result; and adjusting a parameter of a text encoder on the basis of a text recognition result and first supervision data corresponding to the text recognition result, adjusting a parameter of an image encoder on the basis of an image recognition result and second supervision data corresponding to the image recognition result, and a hybrid image-text recognition result and third supervision data corresponding to the hybrid image-text recognition result, and adjusting a parameter of a hybrid encoder on the basis of the hybrid image-text recognition result and the third supervision data corresponding to the hybrid image-text recognition result to obtain the trained network model formed by the text encoder, the image encoder, and the hybrid encoder.

The invention discloses a monocular four-dimensional imaging device and method based on a metasurface. The monocular four-dimensional imaging device comprises an image sensor and an imaging optical system arranged in front of the image sensor. Wherein the imaging optical system comprises a metasurface, the metasurface is designed to simultaneously generate a self-acceleration Airy point spread function with depth-dependent lateral displacement for incident left-hand circularly polarized light and right-hand circularly polarized light, and the self-acceleration Airy point spread function can generate a self-acceleration Airy point spread function with depth-dependent lateral displacement under the condition that a mechanical moving part is not used. Carrying out polarization separation and focusing on the left-hand circularly polarized light and the right-hand circularly polarized light; the image sensor is used for recording a mixed image containing the left-hand circularly polarized light information and the right-hand circularly polarized light information at the same time through single exposure so as to obtain three-dimensional depth and polarization information of the metasurface, namely monocular four-dimensional imaging of the metasurface. The scene perception and understanding capability of a machine is greatly enhanced, and the method has unique advantages in the aspects of target identification, material discrimination, severe environment perception and the like.

The present disclosure proposes methods, apparatuses, computer program products and non-transitory computer-readable medium for providing real-time virtual background in a video session. Real-time environment status information of a target user may be obtained, the real-time environment status information at least comprising geographic location information of the target user. A virtual visual representation corresponding to the real-time environment status information may be determined. A real-time virtual background may be formed through adding the virtual visual representation into a predetermined layout template. A mixed image corresponding to the target user may be formed through combining the real-time virtual background and a real-time human image of the target user. The mixed image may be presented in a user display region corresponding to the target user in a user interface of the video session.

The invention discloses a face in-vivo detection method and system based on scene optimization, a medium and a product, and relates to the field of computer systems based on a specific calculation model, and the method comprises the steps: extracting a mixed image sample, and applying physical drive transformation to obtain an enhanced image sample; scene comparison loss is calculated; based on living body classification loss and scene comparison loss, training the basic model to obtain a teacher model; inputting the scene sample into a scene branch of the teacher model to obtain a scene target vector; inputting the mixed image sample into a scene branch to obtain a scene vector set; calculating the vector similarity between the scene target vector and the vector in the scene vector set, screening out a corresponding image sample, and obtaining a scene matching data subset; taking the teacher model as supervision, executing knowledge distillation, and generating a living body detection model; and inputting a to-be-detected face image to the living body detection model to obtain a classification result. By implementing the method, the living body detection performance of the model in a specific deployment scene can be improved.

Systems and methods for intraoperative tracking and visualization are disclosed. A current minimally invasive surgical (MIS) instrument pose may be determined based on a live intraoperative input video stream comprising a current image frame captured by a MIS camera. In addition, an instrument activation state and at least one parameter value associated with the instrument may also be determined. Intraoperative graphic visualization enhancements may be determined based on the activation state of the instrument, and / or a comparison of parameter values with corresponding parametric thresholds. The visualization enhancements may be applied to a current graphics frame. The current graphics frame may also include visualization enhancements related to proximate anatomical structures with proximity determined from the instrument pose and an anatomical model. The current graphics frame may be blended with the current input image frame to obtain an output blended image frame, which may form part of an output video stream.

The embodiment of the invention provides a medical image segmentation method and device, electronic equipment and a storage medium, and relates to the technical field of image processing. The method comprises the following steps: acquiring original sample medical images of different modalities; determining a dominant mode and a secondary mode; performing image weighted fusion on the original sample medical image and the dominant weight corresponding to the dominant mode and the original sample medical image and the secondary weight corresponding to the secondary mode to obtain a mixed mode image; performing complementary mask processing on the mixed modal image to obtain a mixed image mask; performing feature extraction on each mixed image mask through a pre-constructed initial image feature extractor to obtain initial image features; and performing model training on the initial image feature extractor according to the initial image features and the original sample medical image corresponding to the dominant mode to obtain a target image feature extractor so as to obtain an image segmentation model. According to the embodiment of the invention, the image segmentation precision of the multi-modal medical image can be improved.

The present disclosure provides a method and system for adapting a machine learning model, such as an object detection model, to account for domain shift. The method includes receiving a labeled data elements and target image samples and performing a plurality of model adaptation epochs. Each adaptation epoch includes: predicting for each of the target image samples, using the machine learning model configured by a current set of configuration parameters, a corresponding target class label for the respective target data object included in the target image sample; generating a plurality of labeled mixed data elements that each include: (i) a mixed image sample including a source data object from one of the source image samples and a target data object from one of the target image samples, and (ii) the corresponding source class label for the source data object and the corresponding target class label for the target data object. The method also includes adjusting the current set of configuration parameters to minimize a loss function for the machine learning model for the plurality of mixed data elements. The method results in adapted machine learning model that accounts for domain shift and that has improved performance at inference on new target image samples.

The application discloses an open set image classification field self-adaption method based on self-step learning, first, the original image is preprocessed to obtain an image set, then a feature extraction module and a double multi-class classifier module are constructed and trained to align shared class features of source domain images and target domain images and separate target domain private class features, a multi-criteria cross-domain hybrid module is further constructed and trained, cross-domain hybrid images are generated by using the source domain images and the target domain images, and the shared class features are self-learned, and finally, a classification result of the target domain image is output. Compared with the existing open set image classification field self-adaption method, the application covers smooth and non-smooth class distribution, and does not need to empirically adjust the threshold for distinguishing common class images and private class images in the inference stage, so that the model has good robustness under different hyperparameters and experimental settings.

The invention discloses a quantum classical mixed image classification method based on group isovariant and delayed aggregation. The method comprises the following steps: lifting a convolution module, inputting a target image, initializing a convolution kernel and rotating the convolution kernel, and carrying out convolution and stacking on the target image to obtain an output image; the group convolution module is used for converting the output image of the convolution module into an image sample and carrying out convolution and reverse remodeling operation to obtain an output image; the group pooling and flattening module is used for carrying out adaptive average pooling on an output image of the group convolution module, compressing the output image into a feature vector and obtaining an output feature vector; the group equivariant quantum feature processing module encodes the feature vectors to quantum bits according to the feature vectors output by the group pooling and flattening module; and the delay aggregation and output module measures the quantum bits, aggregates and classifies measurement results, and completes classification of target images. The method provided by the invention has strong nonlinear feature fusion capability, and can realize more accurate image classification.

A device for unmixing images of samples with fluorescent dyes includes one or more hardware processors. The device is configured to obtain a mixed image of a sample; determine an unmixed image based on the mixed image; determine a noise-map image based on the mixed image; determine a signal-to-noise image based on the unmixed image and the noise map-image; determine a denoised signal-to-noise image based on the signal-to-noise image; and determine a noise-reduced unmixed image of the sample based on the denoised signal-to-noise image and on the noise-map image.

This disclosure describes systems, methods, and devices related to video post-processing using a single local directional pattern (LDP) for multiple post-processing steps. A method may include identifying video received by a first device from a second device and decoded by the first device; generating a LDP of a video frame of the decoded video; detecting, based on the LDP and the video frame input to a blurred region detection algorithm, a blurred region and a non-blurred region of the video frame; applying, based on the LDP and the video frame input to a super resolution algorithm, super resolution on the non-blurred region of the video frame without applying the super resolution to the blurred region; and generating, based on the LDP and the video frame input to a blended image algorithm, a blended image of a low-resolution image of the video frame and a high-resolution image of the video frame.

This application provides a multi-scale hybrid DR image enhancement method, apparatus, electronic device, and storage medium. The method includes: acquiring a target image to be identified; dividing the target image into N blocks according to N different block scales to obtain a first block image corresponding to each block scale; for each block scale, performing histogram equalization processing on each first block image corresponding to that block scale according to its corresponding contrast constraint to obtain a corresponding second block image; for each block scale, performing edge smoothing processing on each second block image corresponding to that block scale to obtain a corresponding third block image; and fusing all third block images corresponding to the N different block scales to obtain an enhanced image. This application solves the problem in related technologies where single local histogram equalization is difficult to simultaneously capture image details at different scales.

The invention discloses a single building image reflection artifact elimination method and system based on a multi-scale space attention mechanism, and the method mainly comprises the steps: constructing a single building image reflection artifact elimination network, taking a mixed image as a network input end, and carrying out the reflection artifact elimination of a single building image through the multi-characteristic module coupled single building image reflection artifact elimination network. The output end is a building entity and reflection artifact image; and designing a combined graph perception loss function, constraining the training process of the two-dimensional reflection artifact elimination network of the single building image by utilizing combined convergence of multiple graph perception losses, and guiding the decoupling of the mixed graph layer. According to the method, the reflection artifact elimination problem of the complex building image is solved, and the method is of great significance to image-based intelligent perception and urban ecological inversion.

The invention belongs to the technical field of calculation, and relates to a mixed image data correction system and method, and the method comprises the steps: enabling a first industrial robot to run according to a preset path, continuously shooting the images of a lower road surface in the running process, obtaining the position information and illumination intensity of a marked region in each image, and obtaining the position information and illumination intensity of the marked region; generating standard image information corresponding to the preset path; the first industrial robot sends the standard image information to an upper computer, and the standard image information is synchronized to all the industrial robots through the upper computer; after the industrial robot starts to work, the industrial robot runs according to a preset path, continuously shoots images of a road surface below the industrial robot in the running process, obtains position information and illumination intensity of a marked area in each image, and compares the position information and the illumination intensity with corresponding standard image information; whether driving deviates or not and whether the environment illumination intensity changes or not are judged, so that the calculation power needed by the industrial robot during judgment and comparison is reduced, and the comparison time is shortened.

This disclosure provides systems, methods, and devices for image signal processing that support improved demosaicing of color image signals. In a first aspect, an image processing method includes receiving a first image frame and a second image frame. The method may also include determining a first demosaiced image frame by applying a first demosaicking process to the first image frame, and determining a second demosaiced image frame by applying a second demosaicking process to the first image frame based on the second image frame. A blending weight may be determined based on the first image frame and the second image frame, and a blended image frame may be determined by combining pixel values from corresponding portions of the first demosaiced image frame and the second demosaiced image frame according to the blending weight. Other aspects and features are also claimed and described.

A system for producing an image to design a product can include a processor and a memory. The memory can store a regularizing module, a blending module, a denoising module, and a communications module. The regularizing module can produce a regularized image of a denoised image of an interpolation of a first diffused image and a second diffused image. The regularized image can be regularized with respect to a visual pattern. The blending module can: (1) determine a blending weight and (2) produce, based on the blending weight, a blended image of the denoised image and the regularized image. The denoising module can denoise the blended image to produce the image to design the product. The communications module can cause the image to be sent to a computer-aided design system to design the product.

An operation method performed by an image processor is provided, including the following operations of receiving a mixed image including an input image and a first on-screen-display (OSD) pattern and generating a second OSD pattern, different from the first OSD pattern, based on the mixed image. The second OSD pattern encompasses protection regions of the input image, and the protection regions are in close proximity of the first OSD pattern. The method further includes operations of performing image processing to remaining regions, different from the protection regions, of the input image and generating an output image. The output image includes the second OSD pattern and the remaining regions of the input image, being performed by the image processing.

This invention provides a method, apparatus, terminal device, and medium for hybrid oversampled image enhancement, including: acquiring an original image dataset; constructing a class-reweighted distribution based on the number of effective samples in each class; sampling foreground image samples from the class-reweighted distribution and background image samples from the long-tail distribution; segmenting the foreground and background image samples into non-overlapping image blocks to obtain a set of foreground image blocks and a set of background image blocks; generating a binary mask for each image block; performing element-wise mixing of the foreground and background image block sets to obtain hybrid image samples; extracting neural activation maps of the foreground and background image samples; and using the neural activation maps to assign content-aware soft labels to the hybrid image samples. This invention can improve the recognition accuracy and generalization robustness of Transformer-based visual recognition models for tail categories in long-tail distribution scenarios.

The application discloses a distributed X-ray security inspection machine image judgment system and method. A mode selection unit determines an image judgment mode according to the number of packages passing through an X-ray security inspection machine within a preset time length. In the self-image judgment mode, an image acquisition unit acquires an initial X-ray image sequence, and a local image processing unit processes the initial X-ray image sequence to obtain a pseudo-color X-ray image sequence, and an image judgment display unit obtains an image judgment result. In the hybrid image judgment mode or the external image judgment mode, the image acquisition unit sends a specified initial X-ray image in the initial X-ray image sequence to a cloud image processing unit, the cloud image processing unit processes the received specified initial X-ray image to obtain a pseudo-color X-ray image, and the image judgment display unit obtains the image judgment result. The distributed X-ray security inspection machine image judgment system provided by the application improves the balance of image judgment work, and each X-ray security inspection machine can complete on-demand image judgment, thereby further improving the image judgment efficiency compared with the scheme of centralized personnel image judgment.

The present invention discloses a method and system for identifying abnormalities in electrical components within a meter box. The method comprises: determining whether each infrared image in a set of infrared images is associated with each other; if an infrared image does not have an associated relationship, segmenting the infrared image based on a preset segmentation rule to obtain at least one infrared sub-image corresponding to the infrared image; searching for other infrared images corresponding to the at least one infrared sub-image in the set of infrared images; splicing other visible light images associated with the other infrared images to obtain a target visible light image corresponding to the infrared image; and fusing the target visible light image with the infrared image to obtain a target mixed image. This method effectively expands the application scenarios for identifying abnormalities in meter boxes based on dual spectrums and can improve the accuracy of identifying abnormalities in electrical components.

The invention discloses a tray detection method in an industrial changeable environment based on domain self-adaption, and the method comprises the steps: taking a label-containing tray image data set corresponding to an original tray detection model in an original scene as a source domain data set, and taking a label-free tray image data set in a deployment scene as a target domain data set; performing style migration on the source domain data set and the target domain data set by adopting an image style conversion model to generate a class target domain style data set and a class source domain style data set, randomly intercepting sub-region images from corresponding images in the source domain data set and the class target domain style data set, and splicing the sub-region images into a mixed image to obtain a mixed domain data set; constructing a domain adaptive tray detection model, and training the domain adaptive tray detection model; and calculating a loss function of the student model and the teacher model, and minimizing the loss function to obtain the trained teacher model and student model for tray detection. According to the invention, the problem of domain offset caused by environment change can be effectively relieved.

The application discloses a cross-domain remote sensing image semantic segmentation method based on image mixing and prototype comparison, inputs target domain images into a cross-domain remote sensing image semantic segmentation model, and outputs confidence; and the cross-domain remote sensing image semantic segmentation model is trained as follows: pre-processing source domain images with labels and target domain images without labels; mixing the pre-processed source domain images and target domain images through an adaptive mixing strategy to obtain mixed images; extracting semantic features of the source domain images and the target domain images, and aligning the semantic features of the source domain and the target domain through prototype comparison learning; and training a cross-domain remote sensing image semantic segmentation model based on a semantic segmentation framework Segformer by using the mixed images, the source domain images and the confidence. The application effectively utilizes information interaction between the source domain and the target domain through the mixing strategy, and better helps the model to learn semantic features of the same categories between different domains.

The application discloses a kind of unmanned vehicle real-time ground object segmentation methods based on mixed image, the method includes: collecting the remote sensing image of target area and carrying out partial annotation, making dataset;Ground object segmentation network is constructed and is trained;Real-time ground object segmentation is carried out using ground object segmentation network, the label obtained is visualized and then output to user equipment;The label obtained is carried out confidence calculation, and the label of confidence greater than or equal to threshold value is carried out image mixing processing, generates mixed image, and is input to ground object segmentation network and continues to be trained.The application adopts lightweight neural network based on consistency regularization semi-supervised method, and the delay on mobile device is small, only a small amount of label data is needed, that is, it can be realized, and classification precision is high.In addition, the application adopts image mixing, matching strategy, maximizes the use of data, increases the feature extraction capability of model to low proportion ground object class, improves model segmentation precision, reduces the risk of model overfitting.

This application provides an automated stirring control system and method for the brewing process of baijiu (Chinese liquor). The system acquires material mixing images of the baijiu mixture under red, green, and blue light channels. Based on the material mixing images corresponding to the red and blue light channels, it determines the color difference offset on each intersecting grid in a preset regular grid. Then, based on all the color difference offsets and the material mixing image corresponding to the red light channel, it identifies the edge membership degree of each control grid point in the control grid under the color difference offset state. Multiple edge feature maps extracted based on the texture information entropy of the material mixing image corresponding to the green light channel determine the recognition cost for edge recognition of the baijiu mixture. Based on the recognition cost and each edge membership degree, it performs edge fusion on all material mixing images to obtain the detailed edge features of the baijiu mixture. The technical solution provided by this application can achieve accurate recognition of edge contours in material mixing images under dispersion interference.

The invention relates to the technical field of coal gangue recognition, in particular to a coal gangue classification recognition method and system based on image vision, and the method comprises the steps: obtaining a mixed image of coal and coal gangue through employing a camera array, and dividing a plurality of candidate regions; based on the overlapping position between the candidate areas, the mixing state of the coal and the coal gangue on the overlapping position is determined, and feature loss is used for distinguishing the sheltered area and the unsheltered area under coal gangue mixing; on the basis of continuous frame images of the occlusion area, synchronizing feature loss of the occlusion area to an adjacent non-occlusion area, and determining a spatial similarity threshold value of the adjacent non-occlusion area; performing multi-target cooperative processing according to the spatial position of the sheltered area, and unifying classification labels of coal gangue in each candidate area; and defining region features of each candidate region according to the value range of the classification tag, and taking the region features as synchronously output data. The accuracy and efficiency of coal gangue recognition are realized.

The invention belongs to the technical field of intelligent detection, and particularly relates to an intelligent detection method for hydraulic support side guard plate bending processing, so as to solve the technical problem that the generalization ability of a model is restricted due to limitation of data acquisition and enhancement in the prior art. Cutting a target image from the image A and pasting the target image into the image B to generate a mixed image; s2, performing pixel-level weighted fusion on the mixed image and the image C to generate a mixed image; s3, replacing the corresponding high-frequency sub-band in the mixed image with the disturbance high-frequency sub-band; and S4, pairing the enhanced image with the second mixed label to form an enhanced training sample for training the detection model. The enhanced training sample generated by the method has high fidelity in vision, and is richer in semantic and feature levels.

The invention provides an image processing method, a camera and electronic equipment. The camera comprises a hybrid image sensor and an optical lens, the hybrid image sensor is located on the image side of the optical lens to receive light emitted from the optical lens, the hybrid image sensor comprises first pixels and second pixels, any first pixel can independently obtain light intensity changes and output event signals, and any second pixel can independently obtain light intensity changes and output event signals. The number of color channels of the second pixel is greater than or equal to 7. According to the camera provided by the invention, through the design of the hybrid image sensor, the functions of an event camera and a multispectral camera are integrated into the camera, lens holes of the electronic equipment are reduced, and the overall size design of the electronic equipment can be prevented from being interfered.

The application discloses a line illumination modulation dichromatic tomography imaging system, belonging to the technical field of multicolor tomography imaging, comprising: a line illumination modulation module, comprising two monochromatic light sources and a modulation light path, each light beam is modulated to form a line light spot focused on the focal plane of an objective lens, the position parameter of the line light spot is located on the center of a pixel row or the intersection line of adjacent pixel rows; the distance between the position parameters of the two line light spots is an integer multiple of 0.5 pixels; an imaging module, a multi-element detector with n rows of pixels is used to continuously scan imaging along a first direction, n mixed images are obtained, each mixed image corresponds to a row of pixels, each mixed image contains the signals of 2 channels, n is a positive integer; an image demodulation module is used to subtract the mixed images of two symmetrical pixel rows, the signals of one channel are eliminated from the mixed images, monochromatic images are demodulated, and the monochromatic images contain the signals of another channel. The application realizes synchronous dichromatic tomography imaging on a single detector.