53 results about "Visual contrast" patented technology

The invention relates to the technical field of intelligent control, and discloses an ultra-low power consumption miniature LED visible light intensity self-adaptive regulation and control method and system, and the method comprises the steps: obtaining an analog light intensity signal, and carrying out the smoothing processing of the analog light intensity signal, and obtaining an environment light intensity value; calculating a light intensity change rate, and if the light intensity change rate exceeds a dynamic threshold value, marking an environment state and outputting a light intensity difference value; generating a preliminary adjustment instruction according to the state matching control parameter and the difference value quantization mapping, and calculating target light intensity in combination with the basic brightness; after executing the instruction, calculating actual and target deviation, and iteratively correcting to obtain a light intensity control signal; if the power consumption exceeds the limit, optimizing the signal amplitude; the input driving circuit analyzes the synthesized pulse current, excites pixels to emit light and gathers to obtain adjustment data; and if the visual contrast is lower than a threshold value, the change rate is adjusted again to realize closed-loop refining, and the light intensity adjustment accuracy is improved. The method can improve the real-time sensing capability of ambient light.

A synthetic image device comprises an array of line focusing elements and an image layer (20) having linearly extending cells The image layer is positioned at a distance relative to the array of line focusing elements such that a limited part of the image layer becomes visible when the image layer is viewed through the array of line focusing elements. Each cell comprises at least two sub-cells, provided side-by-side in a transverse direction (T). The sub-cells of the image layer have individual visual-contrast creating items in patterns (21) forming sub-images. Composed images, each associated with a set composed by one sub-image from each cell, are visible from different viewing angles in the transverse direction. Within overlap areas (42E-G), an average pattern density of the visual-contrast creating items in at least two sub-cells giving rise to the overlapping composed images is less than a complete coverage.

The invention discloses a storage battery based on a two-dimensional code traceability function. The storage battery comprises a storage battery body, an exclusive two-dimensional code identifier, a cloud database, a special management APP and an intelligent polishing tool, the exclusive two-dimensional code identifier is printed through acid and alkali resistant UV ink, covers the polycarbonate transparent wear-resistant coating and is permanently fixed on the battery cover of the storage battery body, and the color of the two-dimensional code and the battery cover form obvious visual contrast; the exclusive two-dimensional code identifier is associated with a unique identity identifier, and the unique identity identifier and storage battery full life cycle data are bound and stored in a cloud database. According to the method, the data reading efficiency is improved, the full-life-cycle data association rate is improved, manual one-by-one retrieval is not needed, complete data of the battery from production to operation and maintenance can be directly obtained through code scanning, errors of data recording and querying are reduced, meanwhile, physical coating and encryption dual protection is adopted for the two-dimensional code, the tampering rate is decreased to 0, and the anti-fake capacity is remarkably improved.

The invention relates to the technical field of machining, in particular to a machining method of a support bearing, which comprises the following steps: after finish turning of the support bearing, sequentially carrying out rough grinding of an inner hole, accurate grinding of an outer circle and an end surface, accurate grinding of a large end surface by adopting a surface grinding machine, final accurate grinding of the inner hole and milling of an oil wedge surface. According to the method, the traditional process route of'reference in advance 'is changed, clamping scratches before final accurate grinding of the inner hole are effectively avoided, and the stability of the roughness of the large end face is guaranteed through surface contact machining of a surface grinding machine; and meanwhile, temporary coloring treatment is conducted on the large end face obtained after accurate grinding, a visual contrast layer easy to identify is formed, the problems of inner hole scratches, end face roughness fluctuation and machining reflection interference are effectively solved, and the machining quality, consistency and production efficiency of the supporting bearing are improved.

A synthetic image device (1) comprises an array (10) of line focusing elements (11), an image layer (20) having linear extending cells (22), and a coloured background layer assembly (25). A limited part of the image layer is visible through the array of line focusing elements. Each cell comprises at least three sub-cells (24A-C). The at least three sub-cells having individual visual- contrast creating patterns (21), forming sub-images. Composed images, associated with a set composed by one sub-image from each cell, are visible from different viewing angles. The composed images together present a movement effect. The coloured background layer assembly covers a surface facing away from the image layer. The background layer assembly has a first colouring (28A), visible through the image layer. The first colouring presents a colour varying over the extension of the background layer assembly in registry with movement effects. A producing method is also disclosed.

The invention relates to the technical field of intelligent manufacturing, in particular to an intelligent manufacturing unit-oriented integrated industrial robot control system, which comprises an integrated control center, an initial interaction module, a task scheduling module, a dynamic collaboration module, a collaboration presentation module and an early warning response module, according to the method, identification, matching and testing of robot communication protocols are automatically completed, the equipment access time is remarkably shortened, meanwhile, it is ensured that all robots can stably complete data interaction, system access dead angles caused by protocol incompatibility are avoided, candidate equipment is screened based on production task requirements and the real-time states of the robots, and the production efficiency is improved. Meanwhile, the task execution error probability is reduced through multi-dimensional comprehensive evaluation and in combination with a load capacity deep screening robot, in addition, in the material grabbing link, center deviation is judged through visual comparison, re-grabbing or abnormal early warning is automatically triggered, production stagnation caused by material grabbing errors is reduced, along with reasonable path planning, the production efficiency is improved, and the production cost is reduced. And the robot driving path safety is ensured.

The utility model discloses an electric vehicle hub motor appearance detection device based on a computer vision system, which comprises a conveying mechanism, a carrier, a camera module, a light source module and a computer system, the carrier is connected to the conveying mechanism and can move along the surface of the conveying mechanism, the carrier is used for placing and fixing a hub motor to be detected, and the camera module is connected to the camera module. A rack is fixedly connected above the conveying mechanism, the camera module, the light source module and the computer system are all connected to the rack, the camera module is used for shooting and identifying a hub motor, the light source module is used for illuminating the hub motor, and the computer system is connected with and controls the camera module and the light source module. The camera module can automatically detect and judge the problems of omission, deformation and the like of laser etching characters of products by using an OCR technology, and judges whether the problems of omission of oil seal installation, reverse installation of oil seal, missed installation of screws, valve hole dislocation after assembly of a hub and a side cover and the like exist or not by using a visual comparison technology, so that omission and misjudgment caused by manual inspection are completely eradicated.

This invention relates to the field of welding inspection technology, and specifically to a method and system for detecting welding defects. The method includes: acquiring a depth map of a workpiece weld; performing adaptive pseudo-color encoding on the depth map to obtain a pseudo-color image, wherein a lookup table is constructed based on the cumulative distribution function of the depth map, and depth values ​​are mapped to color values ​​through the lookup table; calculating pixel-level normal vectors based on the depth map, and performing illumination rendering on the pseudo-color image according to the pixel-level normal vectors to obtain a rendered image; inputting the rendered image into a defect detection network, and outputting defect detection results. This invention, by constructing a lookup table based on the cumulative distribution function of the depth map for adaptive pseudo-color encoding, can automatically adjust the color mapping relationship according to the actual depth distribution of each depth map, making different depth values ​​visually present a more uniform and obvious color difference, effectively improving the visual contrast and feature representation of the depth map.

The invention discloses a galaxy form classification method and system based on asymmetric cross-modal momentum contrast learning, and belongs to the technical field of image classification, and the system comprises a visual encoder which extracts image features and stores the image features in a momentum feature queue; the text encoder is used for encoding the text information to generate initial text embedding; the dynamic text embedding projection layer is used for embedding the initial text into the projection to obtain a text projection; the classification head is used for predicting a galaxy form classification result; the processing module is used for calculating total loss including visual contrast loss and cross-modal alignment loss, and updating model parameters of the visual encoder, the dynamic text embedding projection layer and the classification head by using the total loss; for the total loss, the weighted weight of the visual contrast loss is greater than the cross-modal alignment loss. According to the method, asymmetric interaction of images and texts can be realized, text information is effectively fused in a visual dominant galaxy classification task, the classification precision is improved, and resource consumption is optimized.

The invention provides an aim point display method and device, electronic equipment and a storage medium. The method comprises the steps that an aim point graph is displayed on a display picture of an application program in an overlapped mode; acquiring a background image of an area where the aim point graph is located in real time, and extracting color features of the background image; determining a target color according to the color features and a preset color processing strategy, wherein the target color is used for forming visual contrast between the aim point graph and the background image; and updating and displaying the aim point graph by using the target color. According to the scheme, the color of the aim point can be adaptively adjusted along with the change of the background in real time, so that the aim point can be kept clear and distinguishable by forming strong visual contrast with the background in any complex game scene, and the aiming accuracy and the game experience are remarkably improved.

The invention discloses a nickel-cobalt ion extraction visual contrast detection method and an extract liquid detection device, the detection method comprises the following steps: a sampling pump pumps an extract liquid in an extraction tank into a sampling pipeline, a concentration sensor of the sampling pipeline obtains the total concentration of the extracted extract liquid, and the concentration information is transmitted to a terminal; the visual detection camera adjusts the exposure time according to the total concentration of the extracted extraction liquid, photographs and samples the extracted extraction liquid in the sampling pipeline, and then transmits a photographed and sampled picture to the terminal; the color distinguishing module performs color value analysis and decomposition on the RGB color of the extracted extraction liquid picture; and comparing the RGB value of the picture of the extracted extraction liquid with the RGB value of the standard sample of the nickel-cobalt mixed solution under different concentrations to obtain the concentration value of the nickel-cobalt ions in the extracted extraction liquid. According to the detection method provided by the invention, the extraction degree can be accurately and timely fed back, the continuity of the production process is ensured, and the extraction efficiency is improved.

A totally transparent glass jar, intended for containing foodstuffs, with a substantially cylindrical shape; the midsection is characterized by a smooth, continuous and cylindrical surface; the surface delineates a visually neutral area, which enhances, by contrast, the three-dimensional elements located both in the upper and in the lower parts; the underside is characterized by a decorative peripheral band, which consists of four horizontal annular ribs, that extend all around the entire circumference, circularly; a further peripheral decorative band is located symmetrically, at the top, just below the jar neck; this band consists of four horizontal annular ribs, which are identical in number, proportions and arrangement to those of the lower band; all the ribs – both in the upper and in the lower parts - have a purely aesthetic function; the jar neck is narrower than the body and has oblique helical ribs; the open part of the jar is circular and has a regular rim; the individual character of the design arises from: the symmetrical repetition of the decorative bands; the visual contrast between the bands and the smooth cylindrical midsection; and the proportional balance between the body, the decorative bands and the neck; all the reflections and shadows caused by the lights, shown in all reproductions of design No. 1, are disclaimed from the application for registration of design; the No. “22” and the wording “720 ml” shown on the bottom view, are also disclaimed from the application for registration of design; fig. No. 1.1 is a front view of the product; fig. No. 1.2 is a perspective view of the product; fig. No. 1.3 is a further perspective view of the product; fig. No. 1.4 is a bottom view of the product; fig. No. 1.5 is an enlarged view that shows the decorative peripheral band in the underside, which consists of four horizontal annular ribs that extend all around the entire circumference, circularly; fig. No. 1.6 is a top view of the product, which shows its open upper part; fig. No. 1.7 is an enlarged view that shows the decorative band just below the jar neck, which consists of four horizontal annular ribs, identical in number, proportions and arrangement to those of the lower band.

The invention belongs to the field of image processing, and discloses an underwater image quality evaluation method and system based on visual contrastive analysis and brightness-chrominance statistical learning, which enable non-reference quality evaluation to have class reference capability by generating a pseudo original image corresponding to a distorted image to be detected, and remarkably improve the effectiveness of quality difference measurement. By comparing and analyzing the pseudo original image and the distorted image to obtain the first perception quality difference feature, the real perception change of underwater distortion can be described, and the accuracy of perception difference measurement is improved. Second perception quality difference features are calculated based on the similarity of the two, so that the evaluation model can comprehensively consider the structural consistency and the detail degradation degree, and the final quality judgment is more fit for the visual perception rule. According to the method, the accuracy and robustness of no-reference underwater image quality evaluation are greatly improved, and the method is suitable for more engineering application scenes.

This paper describes a double-sided aspherical diffractive multifocal lens and a method for manufacturing and designing such a lens in the field of ophthalmology. The lens includes optics comprising an aspherical front surface and an aspherical rear surface. Multiple concentric diffractive multifocal regions are designed on one of the two surfaces. The other surface may include a toric surface component. Compared to a single-sided aspherical lens, the double-sided aspherical surface design improves the modulation transfer function (MTF) of the lens-eye combination by reducing aberrations and enhancing visual contrast. The surface with multiple concentric diffractive multifocal regions produces near-focal, intermediate-focal, and far-focal points. The phase order of the diffractive annular lens region is shaped into a sinusoidal distribution.

The invention discloses an image classification method and device for mining a semantic relationship by using a large language model, relates to the technical field of image processing, and can improve the recognition and clustering of images of unknown categories. According to the scheme, the method comprises the following steps: acquiring a training sample set, wherein the training sample set comprises a plurality of first images with labeled categories and a plurality of second images with unlabeled categories; for the labeled categories, generating semantic description of each category in multiple dimensions by using a large language model; based on the semantic description of each category, determining semantic relevancy between the semantics of each category and the semantics of other categories to obtain a semantic relevancy matrix; constructing a target semantic visual contrast loss function based on the semantic relevancy matrix, the feature vector of the first image and the temperature coefficient; and embedding the target semantic visual contrast loss function into an original loss function of the GCD task to obtain a target GCD task, and identifying and clustering the images in the training sample set by using the target GCD task.

The application discloses an aerosol generating article and a device used in cooperation with the aerosol generating article. The aerosol generating article is provided with a trigger part. The aerosol generating device comprises a recognition unit. The recognition unit comprises a moving assembly movable between a first position and a second position and a recognition element. The recognition element can recognize whether the moving assembly is in the first position or in the second position. When the aerosol generating article is received in the aerosol generating device, the trigger part makes the moving assembly in a preset position. At this time, the recognition element outputs a recognition signal to make the aerosol generating device heat the aerosol generating article. The preset position is the first position or the second position. Compared with the existing optical recognition or handheld external device recognition, the scheme adopts mechanical movement to trigger the recognition element, does not involve complicated visual comparison or external equipment, is simple to recognize, and reduces the cost of the device.

An edge device can include a processing device, an image sensor, and a memory having instructions that are executable by the processing device for causing the processing device to perform operations. The processing device can receive, from the image sensor, an image of an environment. The processing device can determine a visibility measure corresponding to the environment by determining a dark channel of the image, determining, based on the dark channel of the image, a transmission map of the image, and determining, based on the transmission map, a visual contrast of the image. The processing device can generate information corresponding to the visibility measure.

The application relates to the technical field of intelligent control, and discloses a kind of ultra-low power consumption micro-LED visible light intensity self-adaptive regulation and control method and system, the method includes obtaining analog light intensity signal and obtaining ambient light intensity value by smoothing processing;Calculate the light intensity change rate, if it exceeds dynamic threshold, mark the environment state and output the light intensity difference value;According to state matching control parameter, with difference value quantization mapping generates preliminary adjustment instruction, combines basic brightness to calculate target light intensity;Calculate the actual and target deviation after executing the instruction, and obtain the light intensity control signal by iterative correction;If power consumption is over limit, then optimize signal amplitude;Input drive circuit analyzes and synthesizes pulse current, excites pixel to emit light and converges to obtain adjustment data;Calculate visual contrast, if it is lower than threshold, then re-adjust the change rate to realize closed loop refining, improve light intensity regulation accuracy.This method can improve the real-time sensing ability of ambient light.

The utility model relates to the technical field of interactive equipment, and discloses a Morse code telegraph transmitter, which comprises a transmitter body. The text display area comprises letters and Morse codes; the loudspeaker is arranged in the machine body; the microphone is arranged in the machine body; the microphone button is arranged on the upper end surface of the machine body; the Morse code button is arranged on the upper end surface of the machine body and is pressed down to form a Morse code; the Morse code display lamp is arranged on the upper end surface of the machine body and is used for displaying the Morse code formed by pressing the Morse code button in real time; the control circuit board is used for receiving signals and outputting the signals. According to the utility model, visual contrast is realized, and a user can be trained to be familiar with the Morse password quickly.

In an example, a target includes a first layer having a first surface which includes a target image having a plurality of zones of incapacitation. The zones of incapacitation are highlighted with a visual contrast from a remainder of the target image outside of the plurality of zones of incapacitation. The plurality of zones of incapacitation comprise two or more of a head zone, an upper thoracic zone, a chest zone, or a pelvic girdle zone.

The invention discloses a multi-agent-based single-view simulation scene generation method and device and a medium, and belongs to the technical field of three-dimensional modeling simulation, and the method comprises the steps: recognizing all objects in a single indoor RGB image through a sensing agent, carrying out the positioning and scale calculation of each object, and outputting sensing data; reasoning the spatial relationship of each object in the RGB image through a relationship reasoning agent, and constructing a scene graph; retrieving a three-dimensional model corresponding to each object from the three-dimensional model data set according to each sensing data through a retrieval agent, and performing scale constraint to obtain a three-dimensional model set; placing each three-dimensional model in the three-dimensional model set through a layout intelligent agent according to the scene graph and the pose information of each object; and performing collision detection and visual comparison on the first simulation scene through the optimization intelligent agent, and performing layout optimization according to detection and comparison results. Therefore, by implementing the method, the simulation scene which is accurate in geometry and consistent in semantics can be reconstructed according to the single view.

The invention discloses a visual contrast display box for anti-dazzle and anti-fingerprint pollution effects of optical lenses, which relates to the technical field of display and detection of optical products and comprises a box body, a cleaning and developing device, a wiping device, an optical lens display stand, a mechanical arm, a display panel and a camera. Through the processes of cleaning, wiping, anti-dazzle effect comparison, fingerprint pollution, development effect comparison, cleaning and wiping, visual comparison of the anti-dazzle and anti-fingerprint pollution effects of the optical lens is achieved, a fingerprint development reagent is placed in a cleaning and developing device, visual comparison of the fingerprint pollution degree is achieved, and through image collection and analysis of a light display plate by a camera, the anti-dazzle and anti-fingerprint pollution effects of the optical lens are obtained. The objective quantitative comparison of the anti-dazzle effect is realized, two groups of testing devices with the same conditions are designed in one box body, the anti-dazzle and anti-fingerprint pollution lens and a common lens form a contrast experiment, and the effect contrast visual display effect is achieved.

The invention discloses an injection molding product weld mark detector, and relates to the technical field of injection molding product detection equipment. The device comprises a base, a fixing shell is arranged at the top of the base, driving shafts are fixedly connected to the two sides of the fixing shell, rotating frames are movably connected to the surfaces of the driving shafts, and adjusting rods are fixedly connected to one sides of the rotating frames. A first toothed plate in the longitudinal adjusting assembly is matched with a convex plate, a supporting shaft and a rotating frame are driven to achieve front-back inclination of a workpiece, and a driving shaft is driven to rotate horizontally under the action of a second toothed plate and a baffle in the transverse adjusting assembly; the first visual detection probe and the second visual detection probe can conduct multi-angle surrounding shooting on the top face, the side face and the bottom of a workpiece, the problem that in traditional fixed visual angle detection, side wall defects are likely to be missed is solved, meanwhile, a heater in the reaction assembly preheats the surface of the workpiece and then sprays a detection agent, the visual contrast is enhanced through volatilization differences at weld marks, and the detection accuracy is improved. And efficient and comprehensive detection of the weld marks on the surface of the injection product is realized.

The disclosure relates to a method for micromachining a biological sample for creating a lamella for analysis in a Cryo-Charged Particle Microscope (Cryo-CPM). The method comprising the steps of providing a biological sample on a sample carrier; Locating a sample area on the sample carrier, said sample area comprising a region of interest having biological material from which a lamella can be created; and Micromachining at least part of the biological sample so as to remove material in a part of the sample area surrounding the region of interest, in order to increase a visual contrast between the biological material in the region of interest and its surroundings. With the increased visual contrast a location for a potential lamella can be identified.

The utility model relates to the technical field of household storage, in particular to a cavel cabinet. The cavel cabinet comprises a bottom plate, side plates, a top plate, a back plate, partition plates, a support, screws and a cabinet door. The cavel element is used for reference, the appearance is unique and lovely, the cavel element is distinguished from a traditional cabinet shape, the unique shape design has the distinguishing degree, the storage function is achieved, meanwhile, the cavel element can serve as a space decoration ornament, interestingness and artistic sense can be added to the space, meanwhile, the cabinet face is provided with delicate line decoration, corner treatment is round, the process fineness is embodied, and the appearance is attractive. The whole cabinet is made of wood and leather in a combined mode, the wood shows natural texture, the leather increases soft touch feeling and visual contrast, the texture level of the product is enriched, good novelty is achieved, through the design of joggling, clamping and screw connection, the cabinet body is convenient to assemble and disassemble, installation is easy and convenient, and use is convenient. And a user can assemble conveniently.

The backlight adjustment method and apparatus provided in this application include: when detecting that the current image is a low-brightness scene, dividing the current image into multiple image regions; determining each image dark area from the multiple image regions; wherein each image dark area is an image region with low brightness; and performing the following for each image dark area: obtaining the color value of each pixel in the image dark area; determining the backlight adjustment type and backlight adjustment coefficient of the image dark area based on the color value of each pixel in the image dark area; and adjusting the backlight of the backlight area corresponding to the image dark area according to the backlight adjustment type and backlight adjustment coefficient. The above technical solution can adjust the backlight of the backlight area by dimming or brightening it according to the backlight adjustment type of each image dark area in a low-brightness scene, thereby achieving local adjustment of the backlight to dynamically adjust the visual contrast, retaining more dark details and ensuring compatibility of different scene content in low-brightness scenes.

The invention relates to the technical field of fabric fuzzing and pilling detection, and discloses a fabric fuzzing and pilling detection device based on image processing and a method thereof.The fabric fuzzing and pilling detection device based on image processing comprises an extraction module, a simulation module, a test module, an auxiliary module and an acquisition module, and the extraction module is used for receiving a detection demand text input by a user, accurately confirming the demand text by combining a natural language processing technology with a standard database and adopting a context algorithm, identifying and extracting detection parameters of the demand text, and generating a matched test process. By enhancing the color difference between the pilling and fluffing area and the non-pilling and fluffing area of the fabric, the image shooting effect of the acquisition module is improved, and the auxiliary module has two color development enhancement modes which are based on the acid-base reaction color development and fluorescence selective adsorption principle respectively. And the visual contrast between the fuzzing and pilling area and the original fabric is obviously improved.

The invention discloses an intelligent color matching advertisement typesetting design auxiliary system, and belongs to the technical field of advertisement typesetting design. According to the method, the problems of low efficiency, high cost and uneven quality in the prior art are solved, color and emotion information in images and copywritings can be accurately extracted through image recognition and text analysis, and the color and emotion information is matched with data in a color database; a color matching scheme is scientifically evaluated based on a historical scheme and a design theory, and it is ensured that the scheme meets the standard in the aspects of visual contrast, color proportion and emotion consistency; by providing an interactive typesetting design interface and real-time operation evaluation, a user can intuitively perform typesetting operation and obtain immediate feedback and optimization suggestions, so that the efficiency and the quality of advertisement typesetting design are improved, the professionality and the aesthetic value of the design are improved, and the pertinence and the attraction of the design are ensured.

This invention discloses a weakly supervised speech-video localization method and system based on semantic interaction, belonging to the field of visual-language cross-modal localization. First, this invention uses a pre-trained dataset consisting of speech and corresponding text to pre-train a speech encoder based on four collaborative tasks. Then, it uses a localization task dataset consisting of speech and corresponding video to construct positive and negative sample pairs. Based on the prediction results of the localization network, a binary cross-entropy loss function is calculated. Furthermore, using the candidate segments with the highest scores of the positive sample pairs predicted by the localization network, frame-level positive and negative samples are constructed. Through acoustic-visual contrastive learning, a loss function based on maximizing mutual information is obtained. Combining these two losses, the localization network is weakly trained for localization, achieving efficient cross-modal interaction and improving the accuracy of speech-video localization.