34 results about "Image converter" patented technology

The invention discloses a multi-view graph converter incremental learning method for identifying related diseases. The method comprises the steps of 1, heterogeneous network construction, 2, mask contrast learning, 3, graph converter feature extraction, 4, channel attention fusion, 5, depth matrix decomposition prediction and 6, incremental learning updating. According to the method, a heterogeneous network is constructed to integrate multi-source similarity data, mask contrast learning is utilized to generate multi-view embedding so as to reduce noise, node position information is coded by means of a graph converter, multi-view features are fused through a channel attention mechanism, disease association is further predicted through depth matrix decomposition, and the disease association prediction accuracy is improved. And an incremental learning strategy is designed to realize dynamic updating of new data, a new result is obtained under the condition that original input data and model retraining are not needed, and the method can adapt to the condition of explosive increase of the data.

A structure inspection system includes an image conversion device located in a far edge of the structure inspection system that is proximate to a physical structure, and a near edge processor located in a near edge of the structure inspection system that is remote from the physical structure. The image conversion device receives an image of the physical structure, and renders a real time 3D model of the physical structure based on the first images. The near edge processor is configured to provide a baseline 3D model of the physical structure to the image conversion device. The image conversion device further compares the first real time 3D model with the baseline 3D model, and provides an indication when the first real time 3D model deviates from the baseline 3D model.

A machine learning method according to this embodiment trains a recognizer using a composite image, acquires a labeled real image based on an image captured by a sensor, stores, when at least one of the results of the recognition when the labeled real image is input to the recognizer match the label, results of the recognition performed by the recognizer, performs machine learning using a data set group including a plurality of the data sets, a real image and a composite image forming a pair in each of the data sets, thereby generating an image transformer, generating a labeled real image as a result of the image transformer transforming the composite image into the real image, and trains the recognizer based on the labeled real image.

An image processing apparatus comprises a receiver (401) receiving an image comprising first pixel brightnesses encoded in accordance with a perceptually uniform Electrical to Optical Transfer Function, EOTF. A converter converts this to second pixel brightnesses encoded in accordance with an EOTF representing a logarithmic mapping from optical light values to the second pixel brightnesses and an image processor circuit (411) applies a brightness adjustment processing to the second pixel brightnesses. The converter (410) comprises a first converter (601, 605) generating intermediate pixel brightnesses applying a logarithm function to the first pixel brightnesses A second converter (603) generates intermediate pixel brightnesses by applying a second logarithm function to an output value of the perceptually uniform EOTF for the first pixel brightness divided by a divisor equal to the first pixel brightnesses raised to a power of an exponent having a value exceeding 1. A combiner generates the second pixel brightnesses by combining the intermediate pixel brightnesses.

1. The name of the design product: image instrument. 2. The use of the design product: the design product is used for image measuring instrument, mainly for measuring parts. 3. The design points of the design product: in shape. 4. The picture or photo that can best show the design points: perspective view 1.

Complementary metal-oxide-semiconductor (CMOS) image sensor, comprising: i) a bias junction consisting of a first semiconductor material type and electrically coupled to a voltage source to apply a reverse bias of opposite polarity to a polarity of the first semiconductor material type to the bias junction; ii) a pixel in a pixel array, comprising: a photosensitive element, a first well consisting of a second semiconductor material type, and a second well consisting of the first semiconductor material type, and active CMOS elements used for at least one of 1) reading and 2) resetting the image data acquired by the photosensitive element; and iii) an implanted, buried layer consisting of the second semiconductor material type, extending over a region representing the pixel array below the second well consisting of the first semiconductor material type;wherein the implanted, buried layer, which consists of the second semiconductor material type, is arranged to interact with the blocking bias applied by the voltage source to the bias junction to reduce a parasitic current path between the second well, which consists of the first semiconductor material type and is used by the active CMOS elements, and a substrate, resulting in more uniform pixel-to-pixel performance in an image sensor with two or more pixels.

An image convertor for visual inspection to convert images captured by two visual inspection devices, each of which comprises a main imaging unit that vertically captures images of an inspected object to be subjected to visual inspection, and a plurality of light sources installed between the main imaging unit and the inspected object comprises an image storage unit that stores a first image group and a second image group, a training data set generation unit that generates a pair of images in the first image group and corresponding images in the second image group as a training data set and an image conversion model learning unit that performs machine learning using the generated training data set to generate an image conversion model that converts an image in the first image group into the corresponding image in the second image group.

An image convertor for visual inspection to convert images captured by two visual inspection devices, each of which comprises a main imaging unit that vertically captures images of an inspected object to be subjected to visual inspection, and a plurality of light sources installed between the main imaging unit and the inspected object comprises an image storage unit that stores a first image group and a second image group, a training data set generation unit that generates a pair of images in the first image group and corresponding images in the second image group as a training data set and an image conversion model learning unit that performs machine learning using the generated training data set to generate an image conversion model that converts an image in the first image group into the corresponding image in the second image group.

A construction compliance system includes an image capture device located in a far edge of the construction compliance system that is proximate to a construction project, and a near edge processor located in a near edge of the construction compliance system that is remote from the construction project. The image conversion device receives an image of a construction status of a construction project, and renders a three-dimensional model of the construction status based on the image. The near edge processor receives a three-dimensional model of a construction plan of the construction project, compares the three-dimensional model of the construction status with the three-dimensional model of the construction plan to determine whether the construction status deviates from the construction plan, and provides an indication when the construction status deviates from the construction plan.

An image processing apparatus includes a receiver (401) that receives an image including a first pixel brightness encoded according to a perceptually uniform electro-optical transfer function (EOTF). A converter converts this to a second pixel brightness encoded according to an EOTF representing a logarithmic mapping from an optical light value to the second pixel brightness, and an image processor circuit (411) applies a brightness adjustment process to the second pixel brightness. The converter (410) includes a first converter (601, 605) that generates an intermediate pixel brightness that applies a logarithmic function to the first pixel brightness. A second converter (603) generates an intermediate pixel brightness by applying a second logarithmic function to an output value of the perceptually uniform EOTF for the first pixel brightness divided by a divisor equal to an exponent of the first pixel brightness, the exponent having a value greater than 1. A combiner generates the second pixel brightness by combining the intermediate pixel brightness.

The present invention relates to the field of image conversion technology, and discloses a multi-view conversion method based on a multi-view consistency diffusion model. A multi-view consistency diffusion model is trained through a unified multi-view to multi-view conversion framework based on a diffusion model. The conversion framework has a new self-supervised training strategy, a consistency and adversarial supervision strategy, which utilizes the existing 2D single-view image converter and 3D Gaussian splash technology to generate pseudo-true values ​​as supervision signals, thereby enhancing view consistency and high-frequency details; the multi-view consistency diffusion model trained by the conversion framework also includes a latent multi-view consistency module, which utilizes the latent 3DGS module as the underlying 3D representation to promote information exchange across multiple views, thereby introducing 3D priors into the diffusion model to ensure view consistency. Finally, a joint optimization strategy is proposed, which can achieve a better trade-off between consistency and authenticity.

The present invention belongs to the field of optoelectronic imaging technology. Aiming at the problem that the streak tube imaging lidar developed at present is difficult to meet the detection requirements of large field of view and high precision, a four-dimensional detection system for a large detection field of view and high-precision streak tube imaging lidar is provided, which includes a pulsed laser, a transmitting optical system, a delay system, a receiving optical system, a fiber optic image converter, a streak tube detector, a CCD camera and a signal processing system. Among them, the streak tube detector uses a large detection area streak tube with a single lens focusing system, realizing a large detection field of view and improving the range resolution in three directions. At the same time, it can image under the condition of the movement of the target object, meeting the requirements of large detection range and high-precision diagnosis.

The application discloses a streak image converter tube specific wave band quantum efficiency improving method and a streak image converter tube, and belongs to the technical field of fusion diagnosis and high-energy physics ultrafast diagnosis. In view of the problem that the quantum efficiency of the existing S20 photocathode is low in the blue-green wave band and restricts the detection sensitivity of the streak camera, a medium-metal composite metasurface structure with a subwavelength periodic arrangement is prepared on a cathode incident window, the guided mode resonance effect is used to couple the incident light to a photoelectric emission layer efficiently, and the light absorption and photoelectric conversion efficiency in the target wave band are enhanced. The metasurface structure is composed of a hexagonal grid-shaped distribution circular hole array formed by etching in a silicon dioxide window and an aluminum oxide medium column filled in the hole, and the structure parameters are determined by strict electromagnetic simulation optimization. The quantum efficiency of the cathode adopting the metasurface structure is improved from 14.34% to 19.09% in the 480 nm wave band, the integral sensitivity is improved from 170 mu A / lm to 289 mu A / lm, and the sensitivity of the streak image converter tube in the specific diagnostic window is improved.

The present invention relates to a display apparatus and a method of driving the same. The display apparatus includes a pixel, an image converter that generates a second image by correcting a gray scale of a first logo in a first image of the pixel, and a data driver that provides a data signal corresponding to the second image to the pixel. The image converter detects the first logo based on a value and a saturation of the first image, generates first map data corresponding to the first logo, and specifies the pixel corresponding to the first logo based on the first map data.

The present application provides a method for generating patterns or images on pastries using an AI model, the method comprising at least the following steps: providing a UI module for user to input keywords and descriptions; using an AI-based image module to generate black and white vector images according to the input content; processing the 2D images into digital instructions through an image converter module; sending the digital instructions to a laser patterning module; and using the laser patterning module to apply the patterns to the surface of the cookies.

An image analysis device includes: an image converter configured to generate target image frames based on video image frames captured by a terminal device; a plurality of image analyzers respectively comprising a plurality of image analysis models, the plurality of image analyzers being configured to: generate a plurality of analysis results by analyzing the target image frames respectively using the plurality of image analysis models different from each other, and perform a respective independent analysis scheduling regardless of analysis execution states of other image analyzers; and a security state determiner configured to determine a security mode of the terminal device based on the plurality of analysis results.

Text-to-image transformers configured in one aspect to associate an input text token with the specific object, apply latent blending with attention to a combination of keys and values for the input text token and a background image upon which to add the object; and which in another aspect perform latent blending with attention to keys and values for the object to add, keys and values for the background, and keys and values for a text prompt.

The present invention relates to a display device and a driving method thereof. The display device includes a pixel, an image converter that generates a second image by correcting a gray scale of a logo in a first image for the pixel, and a data driver that provides a data voltage corresponding to the second image to the pixel. The image converter generates first accumulated data by accumulating first map data corresponding to a logo area larger than the logo in the first image during a plurality of frame periods, generates second accumulated data by scaling the first accumulated data per refresh period, generates third accumulated data by initializing values less than a first threshold value in the second accumulated data to a background value, and specifies the pixel corresponding to the logo based on second map data corresponding to the third accumulated data.

The present application relates to the field of astronomical, space science and solar observation coronagraph instrument, especially relates to a kind of external mask type image stabilization coronagraph, it includes in turn: external mask, heat rejection mirror, objective lens group, field stop, field lens, Lyot diaphragm, image converter group and linear polarizer, it further includes: image stabilization device, the image stabilization device includes: image stabilization lens group, guide mirror and image stabilization lens adjusting mechanism, wherein, the image stabilization lens group is located between objective lens group and field stop;The guide mirror is set to the incident light path direction parallel to external mask outside external mask, for collecting solar azimuth and solar elevation angle optical signal, obtains solar pointing deviation signal;The image stabilization lens adjusting mechanism is based on solar pointing deviation signal, adjusts the pose of image stabilization lens group, to compensate solar pointing deviation.The present application combines high-precision imaging and image stabilization technology organically, significantly improves the observation performance of coronagraph, provides reliable technical support for the research of solar activity and coronal structure.

Described herein is a skin detection and prevention system comprising an image converter module for receiving a first photograph and converting the first photograph to one or more new photographs, an image analyzer module for analyzing the one or more new photographs to identify one or more damaged skin areas of the new photographs, and a scoring module for receiving the analyzed second photograph and generating a skin score based on the one or more types of skin damage and the severity present in the one or more new photographs.

The present application provides a self-supervised human pose conversion method and system, and a readable storage medium. The self-supervised human pose conversion method includes: obtaining a human pose skeleton map and a human parsing map; extracting features from the human pose skeleton map and the human parsing map through a pose feature encoder and a decoupled style encoder to obtain human pose features and human part style features; calculating feature correlations through a correlation mining module based on spatial correlation learning; constructing a dense spatial correlation field according to the feature correlations; performing non-rigid deformation on the human part style features based on the dense spatial correlation field to obtain recombined style features; and reconstructing an input human image through an image converter according to the recombined style features. Through the technical solution of the present application, a single human image is reconstructed by extracting and fusing non-aligned features, thereby realizing the supervision process of model training and effectively reducing the costs of data collection and model training.

The invention provides an optical image conversion and transmission device and method based on a single optical fiber, and is used for solving the technical problem that in the existing refractive index modulation type image detection technology, when an image conversion device is far away from a camera, optical path collimation and system debugging are relatively difficult. According to the optical image conversion and transmission device based on the single optical fiber, two-dimensional optical image information can be directly modulated to a spectrum of wide-spectrum laser only through the first optical fiber and the first spatial two-dimensional disperser; the light is converted into emergent light of which the spectrum is in one-to-one correspondence with the two-dimensional spatial position through a second optical fiber and a second spatial two-dimensional disperser, and then the emergent light is collected and recorded by a camera, so that long-distance transmission of the light image is realized. Attenuation of two-dimensional light image information in the optical fiber transmission process can be greatly reduced, and light image transmission is almost not limited by distance.

The present disclosure relates to a machine learning method, a machine learning system, and a computer-readable medium. The machine learning method according to the present embodiment includes training an identifier using a synthetic image, obtaining a labeled actual image based on a captured image captured by a sensor, saving an identification result of the identifier in a case where at least a part of the identification result in a case where the labeled actual image is input to the identifier coincides with a label, generating an image converter by performing machine learning using a dataset group including a plurality of datasets in which an actual image and a synthetic image are paired, converting the synthetic image into the actual image by the image converter, generating a labeled actual image, and training the identifier based on the labeled actual image.

The utility model discloses a compact continuous zoom eyepiece, which comprises an eyepiece sleeve, an eyepiece joint, a zoom hand wheel, a switching circuit board group, a display screen group, an image rotation lens group and an eyepiece group, the eyepiece joint is respectively connected with an external sighting telescope body and an eyepiece sleeve; the display screen group comprises a display screen, a display screen circuit board and a display screen frame; the switching circuit board group is fixed in the eyepiece connector, and the display screen group is fixed in the front end of the relay lens group; the front end and the rear end of the image rotation lens group are fixed to the eyepiece connector and the eyepiece sleeve respectively, the image rotation lens group comprises an image rotation lens tube, a zoom tube, a first focusing lens group, a second focusing lens group and a third focusing lens group, and the zoom hand wheel is arranged on the outer side of the eyepiece sleeve in a sleeving mode and is in transmission connection with the first focusing lens group and the second focusing lens group; the eyepiece group is connected to the rear end of the eyepiece sleeve. According to the eyepiece, zoom adjustment of a traditional white light sighting telescope is fused to a digital sighting telescope, and zoom operation of a user is facilitated.

The utility model discloses a guitar capable of assisting in indicating press positions, which comprises a guitar structure, an indicating structure is fixedly mounted above the guitar structure, the guitar structure comprises a guitar body serving as a body, a placement groove is formed in the surface of the guitar body, a through hole communicated with the placement groove is formed in the lower part of the guitar body, and the through hole is communicated with the placement groove. The indicating structure comprises a single-chip microcomputer, an image converter, an image capturer, a wireless module and a lamp holder module are electrically embedded and installed on the surface of the single-chip microcomputer, the single-chip microcomputer is fixedly installed in a connecting frame, an electronic base is fixedly installed on one side in the connecting frame, and meanwhile the electronic base is electrically connected with the single-chip microcomputer. A button electron for providing a power source is installed in the electronic base in a clamped mode, a frame cover plate is rotatably installed on the upper portion of the surface of the connecting frame, and the guitar capable of assisting in indicating pressing positions can provide a working power source through the arranged button electron.

Augmented reality applications require precise camera calibration to reduce the overall target registration errors. The camera calibration determines the mathematical model of the camera in order to determine how the physical world space is mapped to the camera image space and thus is highly dependent on the optics of the camera. The camera calibration algorithm normally assumes that the optics of the camera is rigidly fixed, which in fact is not at least for tilted laparoscopic cameras as illustrated in FIGS. 6 to 8. Furthermore, any mechanical play between elements of the objective and the image converter will render the optics variable. According to the invention markers to will be placed within the field of view of the laparoscopic camera, which allows for determination of rotation of the optical part and mechanical misalignment without the necessity or providing any additional sensor.

The present invention discloses a differential analysis system and method for traditional Chinese medicine diabetic retinopathy based on artificial intelligence, which relates to the technical field of artificial intelligence and includes a real-time adjustment end, a differential analysis end, and a formula generation end; the real-time adjustment end is used to detect the retinal texture of diabetic patients in real time and automatically generate the retinal image texture information of diabetic patients through an image converter; the differential analysis end is used to receive the retinal texture comparison results of diabetic patients in real time; the formula generation end is used to receive the retinal lesion difference data in real time and based on the standard traditional Chinese medicine formulas set for different lesions according to the retinal lesion difference data. The differential analysis system and method for traditional Chinese medicine diabetic retinopathy based on artificial intelligence can scale or enlarge the retinal comparison image in real time according to the size of the retinas of different patients, improve the accuracy of the differential analysis of diabetic patients' retinas, and improve the accuracy of the differential analysis of diabetic patients' retinas.

Display driving apparatus and driving method. A display driving apparatus for improving sharpness of an image according to an aspect of the present invention includes a luminance calculator for calculating first luminance data corresponding to a first resolution and second luminance data corresponding to a second resolution smaller than the first resolution using input image data, an offset calculator for calculating an offset based on the first luminance data and the second luminance data, an input image converter for converting the input image data into input image data to which the calculated offset has been applied, a first data output unit for generating output image data for a first panel using the converted input image data and outputting the generated output image data, and a second data output unit for generating output luminance data for a second panel using the second luminance data and outputting the generated output luminance data.

An image conversion device includes an image capture device and a renderer. The image capture device captures a plurality of two-dimensional (2D) images. The renderer receives the 2D images and renders a 3D model of an object captured in the 2D images. In rendering the 3D model, the renderer first renders a binary edge map of the object, and next models textures for the 3D model.

The inventive method involves receiving as input a representation of an ordered set of two-dimensional images. The ordered set of two-dimensional images is analyzed to determine at least one first view of an object in at least two dimensions and at least one motion vector. The next step is analyzing the combination of the first view of the object in at least two dimensions, the motion vector, and the ordered set of two-dimensional images to determine at least a second view of the object; generating a three-dimensional representation of the ordered set of two-dimensional images on the basis of at least the first view of the object and the second view of the object. Finally, the method involves providing indicia of the three-dimensional representation as an output.