Patents
Literature
Hiro is an intelligent assistant for R&D personnel, combined with Patent DNA, to facilitate innovative research.
Hiro

30 results about "Image simplification" patented technology

User interface device with touch pad enabling original image to be displayed in reduction within touch-input screen, and input-action processing method and program

A process is disclosed for displaying an image on a display screen and recognizing a user's input by the user's finger touch on a touch screen, the process including: a first step of displaying an original image; a second step of forming a screen pad, such that a reference image, which is a reduced duplicate of the original image, is displayed on the screen pad within the display screen, such that the reference image is overlaid with the original image, in a translucent mode that renders the original image visible through the reference image, or an opaque mode where the original image is not visible. The reference image is obtained by reducing the original image with or without image simplification. A third step, in response to the user touching the screen pad, detects finger position on the screen pad, and recognizes the user's input based on the detected finger position.
Owner:KDDI CORP

Touch screen positioning device and method thereof

The disclosed positioning device for touch pad comprises: a display screen, a DC or image inductor above screen for a given distance and fir to shoot whole screen, a mono-color line light source set at any side of corner of screen whose beam closes to screen as much as possible and covers whole screen; and a high-bright light spot detection module connected between the DC or inductor and computer. This invention simplifies structure and algorithm, and improves anti- disturbance ability.
Owner:GUANGDONG VTRON TECH CO LTD

Image simplification using a robust reconstruction filter

A method simplifies a data structure representing physical measurements of a real-world phenomena, such as an image of a scene of an object. Input data are first acquired, sensed, or measured. If the data are acquired with a camera, then measurement errors for data points in the data structure are not normally distributed. Therefore, an error term for each data point is constructed according to a Lorentzian estimator. The error term can be determined by taking a difference between the value of the data point, and the value at the data point when a model is fitted to the data structure. The error term is then minimized using a downhill simplex minimization process. Finally, each data point is adjusted by the minimized error term to produce a simplified data structure of the real-world phenomena.
Owner:BENHOV GMBH LLC

Photovoltaic cell panel grid line defect detection method and system based on artificial intelligence

The invention provides a photovoltaic cell panel grid line defect detection method and system based on artificial intelligence, and relates to the field of artificial intelligence. The method comprises the following steps: acquiring an image of a single photovoltaic cell panel; constructing an image matrix according to prior color features in the photovoltaic cell panel image, and performing sliding window calculation on the image matrix to obtain a single-channel image; obtaining a compression value of each column of pixels of the single-channel image as a one-dimensional sequence representing the photovoltaic cell panel image; and inputting the one-dimensional sequence into a time convolution network to obtain the periodicity of the grid line in the photovoltaic cell panel image, and judging the defect position of the grid line according to the abnormal point. According to the invention, the semantic segmentation network type is adopted to identify the content of each column in the image, the image processing mode does not need to debug parameters, the periodicity of the content of the photovoltaic cell panel is utilized to adapt to cell panel images of various sizes, the preprocessing process is simplified, and the generalization ability is high. According to the invention, the influence of camera parameters and illumination intensity can be eliminated, automation is realized, and the detection efficiency and accuracy are improved.
Owner:郭燕

Label construction method, system and device and storage medium

The invention discloses a label construction method, system and device for a training model, and a storage medium, and the method comprises the steps: carrying out the label extraction of a figure image, simplifying a complex figure image into a key point two-dimensional coordinate or a two-dimensional mask, and carrying out the training of a generative adversarial neural network model (GAN model); different label images can be generated by simply modifying coordinate positions or two-dimensional mask shapes of key points, figure images corresponding to labels can be generated by inputting thetrained generative adversarial neural network model, and then videos are further synthesized, so that the operation process of figure video synthesis is greatly simplified, and the working efficiencyis improved; subsequently, a new label can be added as a limiting condition according to actual generation requirements, the label and the real image corresponding to the label are sent to the generative adversarial network model for training, and finally, the corresponding real image can be generated through expansion conditions. The method is widely applied to the technical field of image processing.
Owner:RES INST OF TSINGHUA PEARL RIVER DELTA +1

Access control system based on quantum password

The invention discloses an access control system based on quantum password. The access control system comprises an image processing module, an information screening module and a face scanning model arranged on a door body, the output end of the face scanning module is electrically connected with the input end of an image miscellaneous point removing module, and the image miscellaneous point removing module is arranged in the image processing module; the output end of the image miscellaneous point removing module is electrically connected with the input end of an image simplification module, and the output end of the image simplification module is electrically connected with the first input end of an image matching module; and the image matching module is arranged in the information screening module. The access control system utilizes the face scanning module to scan the face of a user, then a mobile phone terminal of the user can receive the corresponding secret key, the user inputs the secret key received by the mobile phone terminal on a secret key input module, so that an electronic lock on the door body is opened, convenience is brought to the user, in addition, a quantum encryption technique is utilized for making the safety property be improved further.
Owner:SHENZHEN POLYTECHNIC

Calibration device and method and depth camera

The invention is suitable for the field of machine vision, and provides a calibration device and method and a depth camera, the device comprises a calibration plate, a moving part, a fixed support, a projection part, a camera and a processor; the calibration plate is provided with feature points and positioning information; the projection part is used for projecting the patterned light beam onto the calibration plate; the moving part is used for changing the position of the calibration plate; the camera and the projection component are fixed on the fixed bracket, and the camera is used for collecting the patterned light beam reflected by the calibration plate and generating a calibration image; and the processor is electrically connected with the moving part and the camera, and is used for receiving the calibration image acquired by the camera, controlling the moving part to move, and determining the internal parameters of the camera and the external parameters of the camera and the projection part based on the calibration image. The calibration images of the calibration plate at different positions are obtained by controlling the moving part, acquisition of the calibration material is simplified, and therefore the calibration process of relevant parameters of the camera is simplified.
Owner:SHENZHEN ORBBEC CO LTD

Face data pair-oriented complete face reticulate pattern restoration method and system and medium

The invention discloses a face data pair-oriented complete face reticulate pattern restoration method and system and a medium. The method comprises the following steps of combindinga face reticulate pattern image and a clean face image corresponding to the face reticulate pattern image into a picture in the horizontal dimension; taking a training set composed of the data pairs as input of a GAN network, and training to obtain a restoration model; and inputting a to-be-restored face reticulate pattern image into the restoration model to obtain a restored clean face image. The method is an end-to-end training mode, the step of reticulate pattern recognition is omitted, and the bottleneck of reticulate pattern segmentation-based restoration is broken through. The image with the reticulate pattern is used as noise data, the face image without the reticulate pattern can be directly generated through end-to-end adversarial training, the integrity of the restored face reticulate pattern imageis improved while the face reticulate pattern restoration process is simplified, and obvious restoration traces cannot appear.
Owner:CENT SOUTH UNIV

Visual perception method and device, computer equipment and storage medium

The invention relates to a visual perception method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring a Bayer format image acquired by an image sensor of a vehicle in a target driving environment; inputting the Bayer format image into an image recognition model to obtain an image recognition result, wherein the image recognition model is a target recognition model obtained by adopting Bayer format sample image training, and the image recognition result comprises the type and the position of a target object in the target driving environment. By adopting the visual perception method, the original format image (namely the Bayer format image) acquired by the image sensor does not need to be converted into the RGB image, so that the processing process of visual perception is simplified, the processing efficiency is improved, and the perception delay is further reduced.
Owner:前海七剑科技(深圳)有限公司

Image generation model and card number identification model training method and device

The invention provides an image generation model and a card number identification model training method and device, and the method comprises the steps of carrying out the processing of a sample three-dimensional card image comprising a first sample card number in a three-dimensional form, and obtaining a sample background image and a sample plane card number image comprising a first sample card number in a plane form; taking the sample background image and the sample plane card number image as a first input image of an image generation model; and determining first target loss according to theoutput image and the sample three-dimensional card image, and adjusting a first parameter corresponding to the image generation model according to the first target loss until the output image of the image generation model is matched with the sample three-dimensional card image to obtain an image generation model comprising the trained first parameter. According to the invention, the three-dimensional card image with the card number in the three-dimensional form can be generated, the requirement for the training sample is simplified, the three-dimensional card image can serve as the training sample to train the card number recognition model, and the recognition accuracy of the card number recognition model is improved.
Owner:BEIJING DIDI INFINITY TECH & DEV

Carbon fiber composite material rivet hole size detection system and method

The invention discloses a carbon fiber composite material rivet hole size detection system and method, which is characterized in that the depth information of the rivet hole contour is scanned by a line laser, and a single rivet hole and a cylindrical hole are respectively obtained by using a deep convolutional neural network algorithm and a projection method image, simplifying the complexity of edge extraction; using the idea of ​​partition, using the Canny algorithm and sub-pixel edge extraction algorithm to more accurately extract the diameter of the cylindrical hole and the large hole on the conical surface; through the two-step method to solve the point cloud data of the conical surface The fitting parameters of the rivet holes are used to realize the angle information of the cone surface of the rivet hole and ensure the accuracy of the angle of the cone angle; the three-dimensional display technology is used to display the scanned three-dimensional image of the rivet hole, which is convenient for intuitive observation of the size information. The method proposed by the invention has the advantages of fast scanning speed, low difficulty of edge detection, high precision of size detection and the like.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

Three-dimensional model attribute image simplification method and device based on rate distortion optimization

The invention discloses a three-dimensional model attribute image simplification method and device based on rate distortion optimization. The method comprises the following steps: acquiring a three-dimensional model attribute image, equally dividing the whole attribute image into a plurality of image blocks, and determining a rate distortion curve of each image block; according to the rate distortion curve of each image block, under the limit of a given total number of pixels, using a Lagrange multiplier method to optimize and solve the down-sampling rate of each image block; and performing down-sampling on the corresponding image block by using bicubic interpolation according to the down-sampling rate of each image block to obtain a simplified attribute image. According to the method, the down-sampling rates of different areas of the attribute image are optimally determined, so that the simplified three-dimensional model with attributes is close to an original model as much as possible in appearance. The optimization process is not only based on the features of the 2D attribute image, but also considers the geometric features of the original and simplified 3D models and parameterization from the model surface to the attribute image. The attribute simplification method provided by the invention has universality.
Owner:北京云境智仿信息技术有限公司

Program image generation method and device and service equipment system

The embodiment of the invention provides a program image generation method and device and a service equipment system. The method comprises the following steps: receiving identification information of a target feature image and a target video transmitted by a WEB client; obtaining the target video according to the identification information; extracting a program image matched with the target feature image from the target video according to the target feature image; and outputting the program image to the WEB client for program editing. According to the embodiment of the invention, the program image is automatically generated according to the video media, the user operation is simplified, and the user experience is greatly improved.
Owner:XIAN NOVASTAR TECH

Mining area ecological restoration guidance system based on big data

In order to solve the defects in the prior art, the invention provides a mining area ecological restoration guidance system based on big data, which comprises an unmanned aerial vehicle patrol flight system and an analysis control device. The unmanned aerial vehicle patrol flight system comprises a patrol flight unmanned aerial vehicle and unmanned aerial vehicle platforms arranged at a plurality of preset positions in a mining area along a patrol flight route of the patrol flight unmanned aerial vehicle. The patrol flight unmanned aerial vehicle at least comprises a high-definition camera shooting towards the ground. The analysis control device comprises an unmanned aerial vehicle flight control module, an image analysis module and an ecological restoration guidance module. The unmanned aerial vehicle flight control module is used for controlling the patrol flight state of the patrol flight unmanned aerial vehicle, the image analysis module receives image data sent back by the patrol flight unmanned aerial vehicle, and a color gamut distribution diagram is formed after image simplification processing. And the ecological restoration guidance module is used for obtaining corresponding ecological restoration guidance information from the big database based on the color gamut distribution diagram and sending the information to the user. On one hand, the working efficiency is improved, and on the other hand, regular monitoring of ecological changes is realized.
Owner:YUNNAN AGRICULTURAL UNIVERSITY

Panoramic picture and video detection method and device, and readable storage medium

PendingCN112766093AAlgorithms to Simplify Face RecognitionImprove accuracyCharacter and pattern recognitionFace detectionImage simplification
The invention discloses a panoramic picture and video detection method and device and a readable storage medium. The panoramic picture detection method comprises the following steps: acquiring a panoramic picture, and presetting an angle parameter; performing angle comparison processing on the panoramic picture according to the angle parameter, and obtaining a parameter comparison result; performing image acquisition processing on the panoramic picture according to a parameter comparison result to obtain a plane picture; performing face detection processing on the plane picture to obtain a detection result, and outputting a face image according to the detection result; wherein the angle parameter at least comprises one of a roll angle parameter, a first overlook angle parameter and a first yaw angle parameter. And performing angle comparison processing on the panoramic picture, and performing image processing on the panoramic picture according to a parameter comparison result to obtain a plane picture. The plane picture is detected through the face detection processing to obtain the face image, the algorithm of face recognition of the panoramic picture is simplified, and the recognition accuracy is improved.
Owner:卓望数码技术(深圳)有限公司

Local image processing method and device, vehicle-mounted system and storage medium

The embodiment of the invention provides a local image processing method and device, a vehicle-mounted system and a storage medium, and the method comprises the steps: displaying a plurality of local image options in a graphical interface; in response to a selection operation on the target local graph option, determining a to-be-displayed image type; generating a to-be-displayed image according to the type of the to-be-displayed image, and displaying the to-be-displayed image in the image interface; wherein the to-be-displayed image is an image displayed at a visual angle of a virtual camera, and a virtual camera coordinate system where the virtual camera is located is a coordinate system constructed by taking the position of the virtual camera under a vehicle body coordinate system as an origin of coordinates. According to the embodiment of the invention, the high-quality local image is displayed in the vehicle-mounted system, the local image can better meet the visual requirements of human eyes through the visual angle of the virtual camera, the local image is reversely mapped to the original image, and the mapping process is simplified.
Owner:HANGZHOU HIKVISION DIGITAL TECH

Screen capture method, terminal device, and computer-readable storage medium

The invention relates to the field of computer technology, and provides a screen capture method, terminal equipment and a computer-readable storage medium, so as to solve the problem of inconvenient operation for users caused by cumbersome operations during the screen capture process. The method includes: when the display screen displays at least two first windows, receiving a screen capture operation input by the user, performing screen capture processing on the screen currently displayed on the display screen, and obtaining at least one target screen capture image, the target screen capture image and the second window The corresponding display screens are in one-to-one correspondence, and the number of target screenshot images is the same as the number of second windows. The target screenshot image is no longer a screenshot including the display screen of each first window displayed, but a corresponding target screenshot image is obtained for the display screen corresponding to each second window, without manually cropping the above screenshot , you can take a screenshot to obtain a target screenshot image corresponding to the display screen corresponding to each second window, which simplifies the operation process and is convenient for the user to operate.
Owner:VIVO MOBILE COMM CO LTD

Panoramic picture face detection method

PendingCN114241579AAlgorithms to Simplify Face RecognitionImprove accuracyCharacter and pattern recognitionFace detectionImage simplification
The invention provides a panoramic picture face detection method. The panoramic picture face detection method comprises the following steps: S1, acquiring a panoramic picture, and presetting an angle parameter; s2, performing angle comparison processing on the panoramic picture according to the angle parameter, and obtaining a parameter comparison result; s3, performing image acquisition processing on the panoramic picture according to a parameter comparison result to obtain a plane picture; and S4, performing face detection processing on the plane picture to obtain a detection structure, and outputting a face image according to a detection result. According to the panoramic picture face detection method provided by the invention, angle comparison processing is performed on the panoramic picture, image processing is performed on the panoramic picture according to the parameter comparison result to obtain the plane picture, and face detection processing is performed on the plane picture to obtain the face image, so that a panoramic picture face recognition algorithm is simplified; and the recognition accuracy is improved.
Owner:ZHEJIANG LISHI TECH

Automatic rain removing method for images shot by vehicle-mounted camera

According to the automatic rain removing method for the vehicle-mounted camera photographed image, raindrops are regarded as noise, firstly, a network model with input and output being images at different moments in the same rainy day scene is established, and rainy day noise distribution is obtained through the network model; and processing the rainy day image through a denoising network to obtain a rain-free effect picture. According to the invention, training is carried out on the basis that input and output are samples of rainy day images at different moments in the same scene, so that rainy day noise distribution is obtained, and rain removal processing is carried out on the rainy day images with the rainy day noise distribution. Thus, according to the invention, the training of the model does not need a clean rain-free image, the data processing difficulty is simplified, the feasibility is improved, and the accuracy degree of data processing is ensured.
Owner:HEFEI NORMAL UNIV +1

Malicious PDF document intelligent detection method and system based on graph structure

The invention discloses a malicious PDF document intelligent detection method and system based on a graph structure. The method comprises the steps of obtaining a to-be-detected PDF document; analyzing a PDF document to obtain a graph structure feature of the document, performing classification importance sorting on all nodes of the graph structure feature, performing retention and merging processing according to a sorting result to obtain a simplified graph structure feature, and calculating a Laplacian matrix of the simplified graph structure feature as an input feature; inputting the input features into a pre-trained convolutional neural network model, and if the output is 1, determining that the document is a malicious document; if the output is 0, judging the document to be a benign document. The graph structure extraction and image simplification method has the advantages that the dimensionality of features is reduced, the training pressure of a deep learning model is relieved, and the efficiency of a system is improved; and according to the input features, the convolutional neural network model is utilized to detect and classify the documents or automatically train parameters of the model, so that the accuracy and usability of the system are improved.
Owner:ARMY ENG UNIV OF PLA

Plant seedling image segmentation method and device, electronic equipment and medium

PendingCN113223016AImprove work efficiencyTimely detection of abnormalities such as improper insertionImage enhancementImage analysisPattern recognitionImage simplification
The invention discloses a plant seedling image segmentation method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring an RGB image of a plant seedling planting area; performing color space conversion on the RGB image to obtain an HSV image of the plant seedling planting area; acquiring a specified alternative mask image according to the HSV image; and segmenting the RGB image according to the specified alternative mask image to obtain a segmented image containing the plant seedling. The RGB image of the plant seedling planting area is spatially converted to obtain the HSV image, the determination factor of the plant color is simplified, the RGB image is segmented based on the specified alternative mask image obtained by the HSV image, the segmented image containing the plant seedling is accurately obtained, the segmentation effect is good, abnormal conditions such as abnormal insertion of the seedling can be found in time, and the working efficiency of plant planting is improved.
Owner:上海西虹桥导航技术有限公司

PCA-based grey image simplification method and apparatus, device and storage medium

The invention provides a PCA-based grey image simplification method. The method comprises the steps of extracting a grey image matrix of grey images; compressing the grey image matrix to obtain a primary compressed grey image matrix; compressing the primary compressed grey image matrix to obtain a secondary compressed grey image matrix; and restoring the secondary compressed grey image matrix to obtain simplified grey images. The invention furthermore provides an active interaction device, a non-transient readable storage medium and an apparatus, which are used for realizing the method. Redundant information of the grey images can be reduced; and spaces occupied in an electronic device by the grey images are reduced.
Owner:CHINA AGRI UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products