12 results about "Image rectification" patented technology

The invention provides an AI human shape recognition perception system and method based on binocular vision, and the method comprises the steps: carrying out the real-time collection through a binocular camera when a doorbell key is triggered, and carrying out the preprocessing of an original image collected in real time; performing coarse parallax estimation on the real-time image rectification to obtain a full-field coarse depth map, determining a human shape candidate region list by using the full-field coarse depth map and combining the heat source region of interest, and performing fine parallax estimation on the human shape candidate region list to obtain a fine depth patch; converting the corresponding fine depth patch into a three-dimensional point cloud set according to the pose information, and performing scale prior screening based on the corresponding three-dimensional point cloud set to obtain a plurality of human shape candidate reserved areas; and performing fusion identification according to the extracted multi-modal features to obtain a human shape identification result. According to the technical scheme provided by the invention, layered parallax estimation and three-dimensional scale prior screening can be carried out on the real-time image to realize high-reliability human shape recognition under the condition of low power consumption, so that the recognition reliability of a sensing system is improved.

Systems and methods directed to image rectification are provided. The method may include capturing, by a pair of imaging devices, a pair of live images of a scene. The method may further include creating a combined image based on the image pair and alignment parameters. The method may further include identifying a first feature in the first image and a second feature in the second image. The method may further include calculating a spatial deviation based on the first feature and the second feature. The method may include identifying whether the spatial deviation exceeds a predetermined threshold, wherein, if so, then the method may include generating rectification parameters based at least on the spatial deviation. Additional systems and methods are also provided.

The invention provides an X-ray back scattering image deformation correction method, device and system, and the method comprises the steps: obtaining the six-degree-of-freedom pose change during the scanning of a device in real time through an environment pose sensing system fixed on the device, replacing the conventional error-prone inertial measurement or contact track recording, and employing the high-precision pose data to achieve the correction of the deformation of an X-ray back scattering image. And calculating the accurate position of each pixel of the original image on an ideal imaging plane through space-time synchronization and an affine transformation model, and finally generating a distortionless image through interpolation resampling. According to the image deformation correction method provided by the invention, high-precision and non-contact full-scene image correction independent of external marks is realized, and the problem of image distortion caused by jitter in handheld scanning is effectively solved.

The invention provides a document image correction method and device based on depth guidance, equipment and a medium, and relates to the technical field of document image correction, and the method comprises the steps: obtaining a color image and a depth image of a to-be-corrected document; splicing the color image and the position prompt image to form a multi-channel input tensor; inputting the multi-channel input tensor into the corrective neural network model to obtain a dense displacement field; wherein the correction neural network model comprises a plurality of cascaded geometric attention blocks, and each geometric attention block is configured to generate a rotation position code and a decomposition attention mask of depth perception based on the depth image, and perform geometric self-attention calculation of row-column decomposition on the feature map by using the code and the mask; and finally, sampling the color image according to the displacement field to generate a corrected document image. Geometric modeling is guided through depth information, a row-column decomposition attention mechanism is adopted, the complex deformation correction precision is remarkably improved, the calculation complexity is greatly reduced, and high efficiency and robustness are achieved.

The application discloses a kind of computer vision and image processing technical field's based on target interception and vanishing point hybrid calculation image rectification method and system, to solve the problem that cannot quickly find vanishing point in the image rectification of prior art, and it is more complex to remove pseudo vanishing point etc., it includes collecting image data, intercepting target image, positioning and linear fitting to target image edge to obtain vanishing point vector data, by vanishing point hybrid calculation model processing to obtain the final predicted vanishing point coordinate, finally carries out image rectification.The application first uses target detection model to detect the position where main target is located, then intercepts, reduces the interference of background information, then the vanishing point hybrid calculation model is fused and calculated to the vanishing point obtained by fitting, to obtain real vanishing point coordinate, reduce vanishing point screening complexity.

The invention discloses a water conveyance tunnel disease detection device and method based on a laser line scanning camera array. The device comprises a mobile detection platform, a laser line scanning camera array, a posture and position measurement unit, a synchronous control unit, a data acquisition and processing unit and a disease identification module. The laser line scanning camera array is annularly arranged in the circumferential direction of the section of the tunnel, and line scanning gray imaging is carried out under active illumination of near-infrared laser. The synchronous control unit controls the trigger frequency and the exposure time based on the minimum recognizable scale of the disease, and motion blur is inhibited; the posture and position measuring unit provides PPS time synchronization and event time marks, and high-precision alignment of images and posture data is achieved; the data acquisition and processing unit completes image correction, splicing and spatial registration to generate a two-dimensional expanded image of the inner wall of the tunnel; the disease identification module automatically identifies crack and leakage diseases based on the image. Full-coverage scanning of the inner wall of the tunnel can be realized under the condition of high-speed continuous movement, and the inspection efficiency and the detection precision are remarkably improved.

The application provides a document image rectification method and device based on depth guidance, equipment and medium, and relates to the technical field of document image rectification, which comprises the following steps: obtaining a color image and a depth image of a document to be rectified; splicing the color image and a position prompt image to form a multi-channel input tensor; inputting the multi-channel input tensor into a rectification neural network model to obtain a dense displacement field; wherein the rectification neural network model comprises a plurality of cascaded geometric attention blocks, each geometric attention block is configured to generate a rotation position code and a depth-aware decomposition attention mask based on the depth image, and perform geometric self-attention calculation of row and column decomposition on a feature map using the code and the mask; and finally sampling the color image according to the displacement field to generate a rectified document image. Through the depth information guiding the geometric modeling and the row and column decomposition attention mechanism, the rectification accuracy of the complex deformation is significantly improved, the calculation complexity is greatly reduced, and the efficiency and robustness are combined.

The application discloses a kind of based on multi-task guide's multi-view feature fusion's distortion document image rectification method and system, comprising: obtaining distorted document image, and distorted document image is input to visual encoder and obtains distorted document image visual feature;Distorted document image visual feature is input two coordinate branches respectively to calculate the world coordinate of distorted image, UV coordinate mapping, and obtain coordinate branch view feature;Distorted document image visual feature is input two plane branches respectively to calculate foreground mask and text line mask, and obtain plane branch view feature;Coordinate branch view feature and plane branch view feature are input multi-view fusion module, and obtain two-dimensional deformation field coordinates;The pixel position of the distortion document image is mapped using two-dimensional deformation field coordinates, and the rectification of distortion document image is realized.The application is suitable for the rectification of most distortion document images with complex background, improves visual effect and text recognition accuracy.

Techniques for image enhancement are disclosed. A method includes receiving an image pair of a scene, the image pair comprising a visible light (VIS) image of the scene captured using a VIS imaging device and an infrared (IR) image of the scene captured using an IR imaging device. The method may further include generating a combined image based on the image pair, wherein the combined image comprises one or more quality characteristics. The method may also include adjusting one or more components of at least a portion of the VIS image based on the one or more quality characteristics of the combined image. The method may also include generating an enhanced combined image based on at least the adjusted VIS image. Additional methods and systems are also provided.

TITLE OF THE INVENTION: PRODUCT WITH A CAPTUREABLE SURFACE IN A RECTIFIABLE IMAGE, METHOD, SYSTEM AND PROGRAM FOR IMAGE RECTIFICATION, METHOD FOR PRODUCING THE SURFACE AND COMPUTER-READABLE MEDIUM. A surface carries a superposition (21) in which first, second and third copies (14, 17, 18) of a two-dimensional visual object are combined. A rectification method for obtaining a rectified image of the surface comprises the steps: - a phase autocorrelation is produced of a first image comprising a portion of the superposition (21), - at least two correlation vectors are identified in the phase autocorrelation, - a transformation is determined that transforms a tuple composed of the correlation vectors into a tuple composed of two known reference vectors for the superposition, and - the transformation is applied to at least a portion of the first image. Figure for the summary: Figure 5

The present application relates to the technical field of image processing. Disclosed are a document image rectification method and apparatus, and a storage medium and a computer device. The method comprises: acquiring a document image to be rectified, and cropping the borders of the document image to be rectified, so as to obtain an initial document image; on the basis of the initial document image, searching a preset standard document library for a standard document image corresponding to the initial document image; by means of a pre-trained document rectification model and on the basis of image features of the standard document image, rectifying image features of the initial document image, so as to obtain image mapping parameters of the initial document image; and on the basis of the image mapping parameters, performing warping processing on the initial document image, so as to obtain a rectified target document image. The method can reduce the training difficulty of a document rectification model and the collection difficulty of sample data and improve the rectification effect of a document image.

The application provides a kind of image rectification method based on binocular disparity, comprising: obtaining the first view image and the second view image of different depth of field of target scene;The first view image and the second view image are adjusted to the same preset size;Determine the relative position attitude of the first view image and the second view image;According to the relative position attitude, the corresponding pixel points in the first view image and the second view image are adjusted to be aligned in horizontal direction;The first view image and the second view image are binocular depth estimation to obtain the first image, and the first view image or the second view image with relatively smaller depth of field is monocular depth estimation to obtain the second image;The first image and the second image are image fusion.The rectification method provided by the application can successfully generate long-range and near view disparity map, and is obviously superior to monocular disparity estimation method in accuracy.