413 results about "Image rotation" patented technology

The invention provides a multi-scale feature fused textile quality detection method, which comprises the following steps of: aiming at multi-scale feature representation, obtaining a torsion angle and a torsion direction feature descriptor which are kept unchanged under image rotation transformation by calculating an invariant moment of a feature vector, and generating a multi-scale torsion feature vector; calculating a correlation coefficient between the torsion frequency and the torsion density in the multi-scale torsion feature vector and the brightness gradient and the brightness change rate in the brightness feature vector, and determining the correlation degree of the torsion feature and the brightness feature; according to the comprehensive feature vector, a support vector machine classifier is adopted to classify the textile image, and whether a defect area with an abnormal torsion area or an abnormal brightness dynamic range exists in the textile is judged; and in the defect area, local features of a torsion angle, a torsion amplitude, average brightness and brightness contrast are extracted, the torsion degree and the brightness abnormal degree of the defect area are judged through threshold comparison, and quantitative description of the defect is obtained.

The invention discloses an achromatic fluorescent laparoscope optical system which is composed of an endoscope set and an eyepiece, the endoscope set is composed of an objective lens with positive focal power and a rod lens set with positive focal power, the objective lens is composed of a first objective lens set and a second objective lens set, and the second objective lens set comprises a first bonding lens and a second bonding lens. The invention has the advantages that: by reasonably arranging the focal power distribution of the lens, the use requirement can be met, the visible light and fluorescence state can be adjusted at any time when the endoscope group is used, additional debugging is not needed, and the allowance is reserved for the rod lens group for subsequent image rotation; meanwhile, through reasonable arrangement of focal power distribution of the lens, design of the balsaming lens and proper control of the length of the objective lens, light is spread gently in the objective lens, the sensitivity of the lens is reduced, the light beam direction can be adjusted, chromatic aberration caused by visible light and fluorescence wave bands can be inhibited, aberration can be corrected, and the imaging quality can be improved.

A method for decoding an image includes receiving information on image partitioning for the image included in a bitstream; obtaining a plurality of image partitions included in the image based on the information on image partitioning; and decoding the plurality of image partitions. Each image partition comprises a plurality of default encoding parts, each default encoding part is selected from a plurality of candidate default encoding parts, each candidate default encoding part is composed of one or more encoding sub-units. The bitstream includes information on rotation of the image.

The present invention discloses a device and method for semiconductor grain double-sided simultaneous fully equal optical path confocal imaging detection based on polarization separation imaging method. The device includes a CMOS or CCD polarization camera, a telecentric imaging lens, a polarization cube beam splitter, a right-angle image rotation prism, a semiconductor grain, and a transparent glass stage for placing the semiconductor grain, which are sequentially arranged in the optical path direction. The advantages of the device and method of the present invention are as follows: It can achieve double-sided simultaneous fully equal optical path confocal polarization imaging detection of the object to be measured, that is, △ = 0, without the need to use a large-depth-of-field telecentric lens to compensate for the optical path difference of double-sided imaging; the double-sided simultaneous polarization imaging detection device has a simple and compact structure and is easy to assemble and debug.

The invention discloses a single-photon three-dimensional imaging system based on double-optical-wedge scanning and a registration method. According to the single-photon three-dimensional imaging system based on double-optical-wedge scanning, multi-beam emission illumination based on Dammann gratings is adopted, so that the energy utilization efficiency can be effectively improved, and the detection distance is increased; by adopting the area array single-photon detector, high-sensitivity detection of extremely weak long-distance target reflected light signals can be realized; an imaging view field can be expanded under the condition that image rotation is not introduced by adopting a double-optical-wedge scanning mode; internal reflection can be effectively inhibited and a close-range detection blind area can be eliminated by adopting sub-aperture common-path coupling; a prior-based rapid reconstruction algorithm is adopted, so that the three-dimensional image reconstruction speed of double-optical-wedge scanning can be effectively improved; by adopting an off-axis focal plane method, high-efficiency coupling registration of emitting laser beams and receiving the field of view of the area array detector can be realized.

The invention belongs to the technical field of remote sensing images, and particularly discloses a remote sensing image rotation high length-width ratio small target detection method and system based on deep learning. Embedding a dynamic refinement rotation convolution module DRRCM with adaptive adjustment capability according to attitude information of a directional target into a backbone network ResNet50 of a target detection model, and extracting basic features; s2, using an anchor point refinement feature alignment module ARFAM to generate a correction prediction anchor frame, and using the correction prediction anchor frame as a guide to dynamically adjust the position of a feature sampling point; s3, training a target detection model by using an IoU threshold method based on high-quality rotation isovariant features generated by DRRCM and ARFAM and the optimized prediction anchor frame; and S4, collecting a to-be-detected remote sensing image, and inputting the to-be-detected remote sensing image into the optimized target detection model to obtain a target detection result. By adopting the technical scheme, the high-quality rotation isovariant feature and the regression prediction anchor frame are generated, and a better detection effect is achieved for detection of small rotating and densely-distributed targets.

The invention relates to the technical field of computer vision and deep learning algorithm optimization, in particular to an image processing method and system based on a fusion growing point optimization algorithm and a Unet network, and the method comprises the following steps: collecting a lumbar vertebra CT image, extracting the gray value and the HU value of each pixel in the image, marking a lumbar vertebra bone region, obtaining the density information of the lumbar vertebra bone region, and obtaining the density information of the lumbar vertebra bone region; analyzing the gradient of the lumbar vertebra, segmenting the initial region, obtaining information points of the basic region from the initial region, classifying, and generating initial structural features of the lumbar vertebra; according to the invention, by collecting the lumbar vertebra CT image and extracting the gray value and the HU value of the pixel, the density information of the lumbar vertebra bone region is obtained and preliminary segmentation is carried out, the gradient change of the region can be accurately analyzed, and a reliable basis is provided for the preliminary region of the image. In the process, the deviation of the lumbar vertebra bone form is successfully corrected by dynamically adjusting the image rotation angle and analyzing the angle difference of the boundary normal, so that the consistency of the form in the subsequent processing process is ensured.

The invention discloses a double-branch contour enhanced industrial product defect detection method based on small samples. The method comprises the following steps: 1, constructing a positive sample image data set and a defect sample image data set of an industrial product; 2, constructing a background removing module, and removing backgrounds of the reference image and the image to be registered; 3, constructing an image rotation coarse registration module, and rotating the first to-be-registered image until the first to-be-registered image is aligned with the first reference image; 4, constructing a double-branch feature-level fine registration module, and solving a mean value and a variance of the feature map c as template features of the industrial product; 5, solving a feature map d and an abnormal score matrix of the same industrial product defect sample image; and 6, establishing an adaptive defect threshold segmentation network to obtain a defect segmentation image. According to the method, better balance between the detection precision and the model universality in a small sample scene is realized.

The invention is suitable for the field of computer vision recognition, and provides a grabbing control method and system of a mechanical arm and electronic device.The method comprises the steps that an image is detected and segmented, and an initial mask area containing non-zero pixel points and corresponding to a detection target is obtained; based on the position information and the color information of the non-zero pixel points, screening target pixel points forming a detection target to obtain a target mask region containing the target pixel points, according to the depth value of the target pixel point, the position information of the center pixel point, camera parameters, a transformation matrix from the target camera to the tail end of the mechanical arm and a transformation matrix from the tail end of the mechanical arm to a mechanical arm base, calculating a target three-dimensional coordinate of the detection target in a mechanical arm base coordinate system; calculating a target grabbing angle based on the image rotation angle of the detection target and a deviation angle between a grabbing tool at the tail end of the mechanical arm and the detection target; and controlling a grabbing tool to grab the detection target according to the obtained coordinates and angles. According to the scheme, the grabbing accuracy of the mechanical arm can be improved.

The utility model belongs to the technical field of semiconductor equipment, and provides a wafer edge defect detection device which is used for wafer edge defect detection and comprises a bearing table top, a first camera shooting part, a rotating shaft and an image processing part, a wafer is placed on the upper surface of the bearing table top, and the first camera shooting part is located above the bearing table top and aligned with the wafer edge; the rotating shaft is connected to the lower surface of the bearing table top and can drive the bearing table top to rotate, and the image processing part is connected to the first camera shooting part and processes the received wafer edge image to generate a wafer profile diagram, and the obtained wafer profile diagram is compared with a wafer template diagram; according to the invention, the edge defect position of the wafer can be obtained, and the edge defect position is marked and stored, so that the detection efficiency and the detection accuracy of the wafer defect are effectively improved.

The invention discloses an image rotation detection and image rotation elimination adjustment method applied to a laser communication optical system, belongs to the technical field of space laser communication, and solves the problem that the space layout compactness, link energy transmittance and the like of the laser communication system are influenced by adopting a traditional rotation elimination device such as a Dove prism and a K mirror. The positions and angles of the left side collimator and the right side collimator are respectively adjusted to realize coaxial and detection reference transmission; placing the detected laser communication system between the two collimators, and adjusting a rotary table pitch angle of the laser communication system to enable a telescope of the laser communication system to be respectively aligned with the two collimators; extracting and recording light spot centroid coordinates of laser emitted by the two light tubes focused on the area-array camera of the laser communication system; calculating to obtain an image rotation angle; and adjusting the angle of a plane mirror in the laser communication system and carrying out iterative detection until the image rotation angle is reduced to meet the design requirement, thereby completing the adjustment work of the image rotation of the pitch axis of the detected laser communication system.

Techniques for implementing image rotation control using reflective waveguide facets in an AR eyewear display system are disclosed. A facet (602) is disposed in a light propagation path located between a first surface (S1) and a second surface (S2) of a waveguide (600) and is arranged to guide propagation of light for a display through the facet from the first surface to the second surface. Using this configuration, light is transmitted through the facet as it guides propagation of the light from the first surface to the second surface, which reduces the amount of image rotations or inversions that would otherwise be induced in the waveguide using conventionally oriented facets and thus simplifies or optimizes the design of other aspects of an AR eyewear display system, such as display source placement or orientation or form-factor bulkiness.

The present invention discloses a device and method for realizing simultaneous equal optical path confocal detection of both sides of a crystal grain by using a polarization image splitting method. The device comprises a CMOS or CCD camera, a telecentric imaging lens, a polarization image splitting prism assembly, a polarization cube beam splitter, a right-angle image rotation prism, a semiconductor crystal grain and a transparent glass stage for holding the semiconductor crystal grain, which are sequentially arranged in the direction of the optical path. The device and method of the present invention have the following advantages: although the double images of adjacent sides of the semiconductor crystal grain are spatially separated on the polarization image splitting prism assembly, the polarization imaging of both sides of the semiconductor crystal grain satisfies complete equal optical path, i.e., △=0, or quasi-equal optical path △≈0.

The utility model provides a laser radar, and relates to the technical field of laser radars, a detection light beam emitted by a transmitting and receiving module is reflected to a target area through a galvanometer module and a reflector in sequence, and the detection light beam reflected by an object in the target area is received by a receiving module; the rotation of the reflecting mirror and the swinging of the galvanometer module are matched with each other, so that the detection light beam rotationally scans the target area, and the driving parameters of the galvanometer module and the driving parameters of the reflecting mirror can be controlled by the control module to be matched with each other in the rotary scanning process; therefore, image rotation compensation is carried out in the process that the detection light beam rotates to scan the target area, so that the position of a light spot projected on the reflector after the detection light beam is reflected by the galvanometer module is unchanged or translated, thereby avoiding the reduction of a vertical view field caused by the influence of image rotation, and enabling the laser radar to scan more areas.

The invention discloses an infrared athermalization super-resolution imaging system and method based on wavefront coding, the system is composed of an imaging main lens group 1, a phase plate 2 and a camera 4, the phase plate 2 is rotated, low-resolution infrared images modulated by phase coding are recorded in the camera 4 in sequence, and a frequency spectrum constraint airspace is adopted to update an iterative algorithm until convergence is achieved. And therefore, the limitation of insufficient Nyquist sampling can be broken through, and the imaging quality of the phase-coded intermediate image can be effectively improved. A phase plate 2 is added on an aperture plane 3 of an infrared imaging system to perform coding regulation and control on an image, so that an infrared focal plane can output a highly consistent intermediate coding image within a certain environment temperature range, and the phase plate is rotated to acquire wavefront coding image information at different angles to realize super-resolution reconstruction of the image, so that the imaging precision is improved. And finally, athermalization super-resolution imaging between-20 DEG C and + 60 DEG C can be realized, and the imaging resolution can be improved by two times.

The invention discloses a vehicle gauge chip integrated packaging manufacturing process based on an etching technology. The vehicle gauge chip integrated packaging manufacturing process comprises the following steps: step 1, designing and preparing a substrate; step 2, preprocessing the chip; step 3, chip mounting and interconnection are carried out; step 4, plastic packaging and molding; 5, heat dissipation and reinforcement are carried out; step 6, image rotation correction and defect detection; step 7, performing post-working procedures and testing; high-density interconnection such as TSV (through silicon vias), RDL (redistribution layer) and micro bumps is realized by using an etching technology, the interconnection density is remarkably improved, the RDL wiring layer manufactured by using the etching technology has the line width less than 2 microns, high-density interconnection on the surface of a chip is realized, the integration level is further improved, the micro-channel is manufactured in the plastic package layer by using the etching technology, and the packaging efficiency is improved. And the copper radiating fins are embedded, so that the radiating efficiency is effectively improved, the problem of local heat accumulation caused by integration of the power chip and the logic chip is solved, a gap between the chip and the radiating cover plate is filled with a thermal interface material with high thermal conductivity, and the thermal resistance is reduced.

The invention relates to an image rotation angle recognition and correction method and system, and relates to the technical field of image processing, and the method comprises the steps: collecting an original image, marking and rotating a corresponding image rotation angle, obtaining a corrected image, and constructing a training set; selecting a corresponding neural network according to the model architecture, and constructing an initial regression model in combination with a preset improved loss function; rotating the corrected image according to a random angle, constructing a training sample, and iteratively training the initial regression model in combination with the training set to obtain a rotation angle recognition model; recognizing and judging the original image through the rotation angle recognition model, and outputting an image rotation angle; rotating the original image according to the image rotation angle to obtain a corrected image; the recognition of any rotation angle of the image is realized by improving the L1 loss function, and the robustness and accuracy of the rotation angle recognition model are improved.

A non-transitory computer-readable storage medium storing an application program which causes an information processing apparatus to perform a control method including acquiring capability information from a printer, determining whether to allow image rotation through a function provided by an operating system of the information processing apparatus based on the capability information acquired from the printer, and providing information based on the determination to the operating system.

The invention discloses a method for automatically processing a portrait photo into a standard certificate photo, belongs to the technical field of computer vision and image processing, and particularly relates to a technology for automatically processing the portrait photo into the certificate photo and a technology capable of automatically adjusting the posture of a figure and enabling the eyes of the figure to be flush with the shoulders. The technology relates to the fields of face detection, key point positioning, image rotation, background synthesis, image segmentation, shoulder key point detection and position adjustment and the like, and is specifically applied to automatic generation of certificate photos. The method for automatically processing the portrait photo into the standard certificate photo is different from traditional manual processing, the portrait photo can be automatically processed into the certificate photo meeting the standard requirement, the figure posture can be automatically adjusted, the eyes of the figure are flush with the shoulders, the complexity of traditional manual processing is avoided, the processing efficiency is improved, and the user experience is improved. Labor is saved, and the processing precision is higher.

The invention provides an anchor-frame-free remote sensing image rotating target detection method based on angle attention, and belongs to the technical field of computer vision and remote sensing image processing. The method comprises the following steps: inputting a remote sensing image and extracting multi-scale features by using a backbone network; an optimal angle feature is extracted and selected through an angle feature extraction module and a feature pyramid network; a detection head is added to multi-scale output of the feature pyramid network for classification, frame regression and angle parameter regression; a frequency domain channel attention mechanism is introduced into an angle regression branch through an angle prediction head module in a detection head, so that the angle of a rotating target is predicted more accurately; and carrying out joint training by utilizing ProbIoU loss and distributed focus loss to realize end-to-end high-precision rotating target detection. According to the method, relatively high detection precision and reasoning speed are realized on a plurality of public remote sensing image data sets, and the method has efficient performance.

The invention discloses an eight-fold-ratio white light sighting telescope which comprises an objective lens group, an eyepiece group, an image rotation lens group, a zoom hand wheel and a telescope tube. The objective lens group is arranged in the front end of the lens tube; the image rotating lens group is arranged in the rear end of the lens tube; the zoom hand wheel is connected with the steering lens group and is used for adjusting the axial position of an optical element in the steering lens group; the eyepiece group is connected to the rear end of the lens tube. The sighting telescope can achieve multi-magnification adjustment, and is simple and stable in structure and easy to use.

The invention provides a small sample image classification method based on self-supervised task aided training, and the method comprises the steps: carrying out the geometric rotation transformation of a support set image in a training stage, generating an extension set, and assisting a conventional scene training task through constructing an instance-level image rotation angle judgment task and a set-level classification consistency task. The measurement network model comprises a feature extraction module and a measurement module, and a three-branch loss optimization path is constructed: a top-layer branch executes traditional scene training to minimize classification loss; the intermediate branch constructs an instance-level proxy task by predicting an image rotation angle; and the bottom layer branch performs classification prediction on different extension sets to construct set-level classification consistency loss. Through joint optimization of a multi-branch loss function, model learning is guided to be independent of feature representation of category labels, and the generalization ability of the model is improved. According to the method, the deep structure features of the image can be effectively captured, and the classification precision and the model generalization ability are synchronously improved in a data scarce scene.

Features described herein generally relate to a graphical user interface that includes an insight hub and an insight carousel. In one example, a system can receive insights associated with transactions, generate an image carousel that includes slides populated with information associated with the insights, and generate another interface page that includes insight listings corresponding to multiple insights.

The invention discloses a flange plate metal piece surface flatness detection method and system, and relates to the technical field of flatness measurement. According to the flange plate metal part surface flatness detection method, a current detection image of a flange plate metal part is obtained based on an industrial camera; acquiring a comparison reference image from a reference image group based on the current detection image; based on the comparison reference image, rotating the current detection image to obtain a feature alignment image; and based on the comparison reference image and the feature alignment image, acquiring a defect feature value of the flange plate metal part in the current detection image. According to the method, the detection images corresponding to the flange plate metal piece with the qualified surface are selected from the historical detection images to form the reference image group, and then the reference image which is most similar to the current detection image in illumination condition and shooting angle is screened from the reference image group to be compared with the current detection image; and further accurately judging whether the surface of the flange plate metal part in the current detection image has defects or not.

The invention discloses a remote sensing image rotating target detection method, and belongs to the field of high-resolution remote sensing image processing and rotating target detection, and the method comprises the following steps: extracting multi-scale features of a target remote sensing image to be detected; inputting the multi-scale features into a hybrid Mamba feature pyramid network to obtain enhanced multi-scale features; based on the enhanced multi-scale features, RoIAlign is adopted to generate rotation candidate region features; and inputting the rotation candidate region features into a decoupling directed detection head for classification and discrimination to obtain the category and spatial position information of the target.

The invention discloses a rotating target detection method based on frequency division feature refinement. The method comprises the following steps: firstly, carrying out initial feature extraction on an input image; then, carrying out frequency division feature refinement on local features, global features and context features of four stages, paying attention to image key position information, modeling a global relationship, extracting surrounding environment information as supplement, and enhancing the capturing capability of the model on the information; and finally, classifying and positioning the target by using a detection head, designing a sample collaborative optimization loss, adaptively adjusting the attention degree of the model on positive and negative samples, and further improving the detection precision of the model. The method solves the problems that in the past, remote sensing image rotating target detection feature extraction capacity is insufficient, interference and shielding between targets are caused by complex scenes, and target features are sensitive to positioning. The method is suitable for a rotating target detection task of a remote sensing image in a complex environment, and is high in detection accuracy and good in robustness. The method is widely applied to the field of rotating target detection.

The invention provides a remote sensing image rotating target detection rapid deployment method, system and device. According to the method, a remote sensing image target detection model is efficiently mapped to a DPU (Data Processing Unit) for execution through a Vitis AI quantification tool and an executable file, so that the computing resources of an FPGA (Field Programmable Gate Array) are utilized to the maximum extent, and the computing efficiency is improved; in the mapping period, operators which are not supported by the DPU are optimized and equivalently replaced, it is ensured that the preset target detection model can all run on the DPU, and effective distribution of the operators is achieved; secondly, an ONNX model format and a Viis AI interface are adopted, so that the compatibility and expandability of the preset target detection model are enhanced, and the preset target detection model can be seamlessly migrated and deployed on different frames and hardware platforms; and finally, the deployment process of the preset target detection model on hardware is simplified, the technical threshold and deployment cost of developers are reduced, and meanwhile, the real-time performance of the system is improved.

The invention provides a technology for improving visibility of a projected image in a portable information terminal capable of adding an image projection function. A portable information terminal includes a terminal body and an accessory mounted on the terminal body and communicably connected with the terminal body. The attachment has a projection unit that projects an image on the basis of image information obtained from the terminal body, and is configured such that the direction of a projection opening of the projection unit is variable with respect to the terminal body. The terminal body or the accessory determines a state of the portable information terminal including a position of the accessory relative to the terminal body. The terminal body or the accessory performs projection control of the image based on the determined state. The projection control includes, for example, image rotation control, image shape control, or projection angle control.

The present invention relates to the technical field of computer vision and deep learning algorithm optimization, specifically to an image processing method and system based on a fusion growth point optimization algorithm and a Unet network, comprising the following steps: collecting lumbar CT images, extracting the grayscale value and HU value of each pixel in the image, marking the lumbar vertebrae region to obtain density information of the lumbar vertebrae region, analyzing the lumbar vertebrae gradient and segmenting the initial region, obtaining basic region information points therefrom and classifying them, and generating preliminary structural features of the lumbar vertebrae. The present invention, by collecting lumbar CT images and extracting the grayscale value and HU value of pixels, obtains density information of the lumbar vertebrae region and performs preliminary segmentation, can accurately analyze the gradient changes in the region, and provides a reliable basis for the preliminary image region. In this process, by dynamically adjusting the image rotation angle and analyzing the difference in the boundary normal angle, the deviation of the lumbar vertebrae morphology is successfully corrected, thereby ensuring the consistency of the morphology in the subsequent processing process.

The embodiment of the present application provides a method, device, storage medium and program product for executing a rotation operator, which relates to the field of artificial intelligence technology. The method includes: dividing the image to be processed in the video memory into multiple original pixel blocks, each original pixel block includes multiple pixel points, and then loading the entire original pixel block into the on-chip cache, which increases the granularity of data transmission between the on-chip and off-chip, reduces the frequency of accessing the video memory, and thus reduces the memory access pressure. The pixel points in the original pixel block are read from the on-chip cache to perform the rotation operation, and the pixel points are written out to the on-chip cache according to the target coordinates after rotation. When the rotation operation is completed for the entire original pixel block, the target pixel block obtained by rotation is written out from the on-chip cache to the video memory. In other words, the present application uses the on-chip cache to buffer the data and uses the on-chip high bandwidth to perform the image rotation operation, thereby improving the performance of the image rotation operator.