18results about How to "Reduce blur" patented technology

The invention discloses a complex-material-oriented multi-path full-frequency global illumination neural rendering method and system, and belongs to the technical field of computer graphics, and the method comprises the steps: carrying out the scene query of light emitted by pixels, and obtaining a primary geometric buffer and a multi-path mixed geometric buffer of the multi-time ejection of the light; respectively extracting main frequency features and multi-path sub-frequency features in a neural feature field based on the two features; performing directed graph structured modeling based on the multi-path sub-frequency features to obtain propagation features representing light path propagation information and global features representing overall light path information; performing multi-path frequency joint feature fusion based on the dominant frequency feature, the propagation feature and the global feature to generate a fusion feature with a full-frequency feature; and estimating a final radiation brightness value of the pixel based on the fusion feature, and completing neural rendering of global illumination. According to the method, efficient modeling and high-quality drawing of multi-path full-frequency global illumination under a complex material can be realized, and the sense of reality and dynamic consistency of a rendered image are remarkably improved.

The application provides a cloud image efficient synthesis method, belonging to the field of image processing, comprising the following steps: calculating the edge distance transform of the mask of the foreground image to generate a multi-level boundary information graph; converting the background image and the foreground image to the Lab color space, and inputting the Lab channel, the mask and the multi-level boundary information graph into a generator network to output a preliminary fusion image in the Lab color space; using a brightness discriminator and a chroma discriminator to respectively judge the local texture consistency of the preliminary fusion image and a real fusion image in the L channel, and the global color harmony of the preliminary fusion image and the real fusion image in the a and b channels; performing the adversarial training of the generator network and the discriminator based on a total loss function until the network model converges, and using the converged generator network to generate a final fusion image. The application can guarantee the high realism and consistency of the fusion image in the details and the overall color tone.

The utility model discloses a guide rail type focusing device, and relates to the technical field of guide rail type focusing. The guide rail type focusing device comprises a transmission gear, a fixing support, an image sensor and a stepping motor. The transmission gear is in sliding connection with the fixing support, the image sensor is arranged on the fixing support, a spring is further arranged between the image sensor and the fixing support, a driving gear is arranged at the output end of the stepping motor, and the driving gear is connected with the transmission gear in a meshed mode. According to the guide rail type focusing device provided by the utility model, the image sensor is arranged on the fixed bracket, and the fixed bracket performs lifting motion so as to achieve the purpose of focusing. In addition, a guide rail mode formed by the transmission gear and the fixing support enables a guide rail inclined groove and an inclined bone position to be in surface-to-surface contact so as to achieve the purposes of stability and zero clearance, assembly tolerance is reduced, focusing precision is improved, high-speed snapshot is achieved, meanwhile, the whole image is clear and balanced, and the problem that one side of the image is clear and the other side of the image is blurred is solved.

The invention discloses a network camera data storage system and method based on wireless transmission, and belongs to the technical field of wireless transmission, and the system comprises a wireless link sensing module which is used for monitoring original parameters of each wireless transmission path in real time, calculating the original parameters, generating a real-time quality score of each wireless transmission path, and sending the real-time quality score to a server; outputting a link quality spectrum; the data processing module is used for acquiring original video data acquired by the network camera and outputting coded video data; the energy consumption decision module is used for generating an equipment working mode instruction; and the path execution module is used for selecting a main transmission path and an auxiliary transmission path from the link quality spectrum and transmitting the coded video data to a preset rear-end storage system through the selected path, and the method comprises the steps A1-A5. The main transmission path and the auxiliary transmission path are intelligently selected through the link quality spectrum, the long-term working stability of the system in a complex wireless environment is improved, and the service life of the system in the complex wireless environment is prolonged.

This invention belongs to the field of image processing technology, specifically relating to an online monitoring method and system for the drying morphology of instant goat milk powder. The method includes: acquiring images of the fluidized bed interior to construct a spatiotemporal fluidized image tensor; obtaining fluidization confidence using eigenvalue decomposition of the local spatiotemporal structure tensor matrix to determine the effective fluidized region; performing morphological opening operations on the effective fluidized region using a sequence of structuring elements, statistically summing gray values ​​and constructing a morphological pattern spectrum to obtain the morphological pattern spectrum entropy; obtaining the pulsating velocity vector using an optical flow algorithm, thereby obtaining the average particle pseudo-temperature and coefficient of variation; and obtaining the fluidization steady-state potential energy index based on the morphological pattern spectrum entropy, average particle pseudo-temperature, and coefficient of variation, and generating control commands. This invention achieves monitoring of complex and abnormal conditions such as the initial stage of dead bed formation and channeling through the fluidization steady-state potential energy index.

The novel zoom indoor spotlight comprises a shell, a light source assembly, a focusing assembly and a lens assembly, and the focusing assembly comprises a rear lens, a front lens and a focusing mechanism; the small end of the rear lens is close to the light source assembly, and the surface of the small end is provided with a light inlet hole. A shaft hole is formed in the middle of the large-end surface of the rear lens, and a plurality of first annular concave-convex surfaces in a concentric ring array are arranged on the large-end surface; a rotating shaft rotationally matched with the shaft hole is arranged in the middle of the back surface of the front lens, and a plurality of second annular concave-convex surfaces arrayed by concentric rings are arranged at the non-middle position of the back surface; a plurality of first convex parts which are distributed circumferentially are arranged on the first annular concave-convex surface, and the surface of each first convex part is an arc-shaped convex surface; a first concave part is arranged at the position, corresponding to the first convex part, of the second annular concave-convex surface, and the bottom surface of the first concave part is an arc-shaped concave surface. The utility model has the characteristics of stable and accurate focusing, sealed and durable structure, small volume, wide application range, long service life and the like.

The application provides a kind of underwater image recognition method based on bayesian neural network, belong to underwater image recognition technical field;Solve the technical problem that general target recognition method is poor in robustness in visual degradation underwater environment.The method steps include: S1, constructing underwater submarine image dataset;S2, construct underwater image recognition model based on bayesian neural network;S3, use the constructed dataset to train deep learning model, obtain underwater image recognition model;S4, according to the trained prediction model, the image collected when executing underwater task to submersible is predicted and analyzed, and the image target classification result and uncertainty are obtained.

The application discloses an environment sensing method and device for rail transit, and the method comprises the following steps: acquiring continuous frame image information F n ; performing Gaussian filtering processing on the continuous frame image information F n to obtain initial processing image information; performing fusion operation on the initial processing image information to obtain fused image information; performing reprocessing operation on the fused image information to obtain reprocessing image information; and determining environment sensing information of rail transit based on the reprocessing image information. Through data optimization processing and detail enhancement processing on the continuous frame image collected in the rail transit process, the data analysis and processing amount in the environment sensing process is effectively reduced, the environment sensing efficiency is improved, and the accuracy of environment sensing is effectively improved.

This utility model discloses a drying component for an intelligent water-based ink printing machine, belonging to the field of printing machine technology. The drying component includes a mounting box with multiple support blocks fixedly connected to its bottom. The rear end of the mounting box has an inlet and an outlet. An installation groove is provided inside the mounting box, on which a drying component and a tensioning component are installed. The drying component dries the printed paper, and the tensioning component adjusts the tension of the printed paper. By setting up the drying component, multiple electric heating tubes operate simultaneously, rapidly increasing the temperature inside the mounting box and causing the moisture in the ink on the surface of the printed paper to evaporate quickly. Furthermore, hot air is sprayed from the jet pipe for secondary drying, further accelerating the drying process and significantly shortening the drying time. The tensioning component reduces paper movement and improves print quality.

The utility model discloses a body limiting structure for infant magnetic resonance, which comprises a body supporting rod, limiting supports are fixed on the outer side of the body supporting rod, the body supporting rod and the limiting supports are symmetrically arranged, the two symmetrical limiting supports are connected, a limiting concave part is arranged at the joint, a connecting rod is fixed at one end of the body supporting rod, and the other end of the body supporting rod is connected with the limiting concave part. One end of the connecting rod is connected with the body supporting rod, the other end of the connecting rod is internally and rotatably connected with the arm supporting rod in an inserted mode, a fixing structure is arranged between the body supporting rod and the arm supporting rod, arm soft cushions are fixed to the outer sides of rod bodies of the arm supporting rod, and adjustable binding bands are sewn to the surfaces of the arm soft cushions. By adding the structural design of the arms, the difficulty that an adult assists an infant in keeping a specific posture of raising the arms is reduced, the operation complexity of medical staff is reduced, and the working efficiency is improved. Due to the design of the arm cushions and the adjustable bandages, the arms of the infants can be placed more comfortably, and the discomfort of the infants in the examination process is reduced.

The application discloses an edge enhancement and bottleneck vector-based referential image segmentation method and system, relates to the technical field of referential image segmentation, and comprises the following steps: training an incremental cross-modal bottleneck fusion network based on a plurality of sample images and text descriptions of each sample image; in the incremental cross-modal bottleneck fusion network, embedding and representing the text description of a preset sample image through a text encoder; in the process of multi-stage visual extraction of the preset sample image by a visual encoder, introducing the embedding and representation result into the visual features at each visual extraction stage by using a cross-modal attention fusion module; in the process of multi-stage decoding of all visual features by a decoder, generating local edge enhanced features at multiple decoding stages by using a local edge refinement module; and the incremental cross-modal bottleneck fusion network trained can accurately perform referential image segmentation on a target image.

This invention discloses a low-light image enhancement method based on random region concealment and reconstruction. The network consists of two stages. In the first stage, the network is trained on a well-lit image dataset to reconstruct the noisy image after random regions are concealed, resulting in a denoising and reconstruction feature encoder for low-light noisy images. In the second stage, the network inherits the feature encoder model parameters from the first stage and inputs the extracted different features into a multi-scale feature fusion module guided by the reconstructed features. This allows the network to retain more image detail features while denoising. This invention utilizes a richer dataset of well-lit images, saving time and manpower by capturing a large number of real paired low-light images for network training. By fusing features extracted from different feature encoders, the network can quickly adapt to real low-light images under complex lighting conditions, achieving good low-light enhancement and denoising effects.

This invention relates to the field of geometry and drift correction technology for cone-beam CT systems, specifically providing a device and method for synchronously correcting the geometric position and focus drift of a cone-beam CT system. The device includes a correction phantom, a radiation source unit, a moving unit, and a detection unit. The phantom consists of a transparent pressure plate and positioning steel balls, positioning steel bars, and a cross steel bar sandwiched between them. The positioning steel balls are arranged in a square area, with a pair of vertical sides arranged at equal and non-equal intervals to provide angular identification features. The positioning steel bars are located in the middle of the horizontal sides, and the cross steel bar is located at the center of the area, with the intersection point serving as the drift measurement reference. This invention, through its integrated phantom design, can simultaneously complete geometric parameter calibration and focus drift measurement in a single scan without moving or replacing the phantom. The correction method integrates algorithms such as elliptical trajectory fitting, feature point matching, and bicubic spline interpolation, enabling efficient and accurate calculation of the dynamic curves of detector position, attitude, and focus drift.

The utility model relates to the technical field of memory pillow processing, and discloses a memory pillow textile fabric surface pattern pyrography device which comprises a working table, a supporting frame is fixedly installed at the top of the working table, a controller is arranged on one side of the supporting frame, sliding grooves are symmetrically formed in the top of the working table, and the sliding grooves are communicated with the controller. A clamping assembly is arranged at the top of the workbench, and a pyrography assembly is arranged at the bottom of the supporting frame. Fabric is flatly laid on the surface of an object placing plate, a rotating block is rotated, a threaded rod rotates and moves downwards, a movable plate and a pressing plate are driven to press downwards, the fabric is fixed to the object placing plate, a short guide rod ensures that the pressing plate vertically presses downwards, and deviation is avoided; the movable rod drives the clamping block to be inserted into the clamping groove of the movable plate, the spring resets to provide pulling force, the pressure of the pressing plate is locked, and the effect of avoiding pattern deviation caused by fabric displacement is achieved.

The invention relates to a nuclear medicine image generation method and device, computer equipment, a storage medium and a program product, which are applied to the field of image processing, and the method comprises the following steps: obtaining motion information and scanning raw data of a scanned object at each time point during scanning; determining a three-dimensional coordinate of the scanned object at each time point according to the motion information of each time point; according to the three-dimensional coordinates of the time points, merging the motion information of the time points to obtain a plurality of sets; and obtaining a nuclear medicine image of the scanned object according to the motion information in each set and the scanning raw data. By adopting the method, the calculation efficiency can be improved while the motion correction effect is ensured.

This invention discloses a motion compensation method for UAV-borne Mini-SAR imaging based on a multi-scale state-space network. Addressing the image defocusing problem caused by residual azimuth phase error in UAV Mini-SAR, this invention constructs a dual-channel input image block from the real and imaginary parts of the complex-valued SAR image with motion error. It extracts defocusing features at different scales using a multi-scale large-kernel convolution module and establishes a long-range dependency along the azimuth direction using an azimuth state-space enhancement module, outputting polynomial phase error coefficients. Subsequently, a phase compensation term is constructed based on these coefficients, and explicit phase correction is performed in the azimuth frequency domain. The focused image is then obtained through inverse transformation. This invention also discloses a motion compensation system for UAV-borne Mini-SAR imaging based on a multi-scale state-space network, used to execute the compensation method of the previous embodiment. This invention can achieve effective estimation and rapid compensation of residual azimuth phase error, possessing physical interpretability, high compensation accuracy, and high processing efficiency.

This utility model relates to the field of packaging box printing technology, specifically to a gift packaging box printing device, including a base plate, a support plate, and a side plate, as well as a controller, a printing mechanism, a flattening mechanism, and a blower mechanism. The printing mechanism includes a screen printing plate, a printing frame, a guide assembly, and a pushing assembly. The flattening mechanism includes a lifting assembly, a connecting assembly, and two pressure rods. The blower mechanism includes a position sensor and a rotating assembly. The pushing assembly includes a first motor, a first connecting rod, a second connecting rod, a connecting block, a first scraper, and a second scraper. The connecting assembly includes two rotating plates, two sliders, and four rollers. The rotating assembly includes a second motor, several fans, several hinge shafts, several synchronous pulleys, and several synchronous belts. This gift packaging box printing device achieves flattening of the packaging box before printing and drying of the printing ink on the packaging box after printing, thereby improving the printing effect and printing quality.

The utility model relates to the technical field of medical instruments, and discloses a pneumoperitoneum connecting pipe with a filter structure, which comprises a connecting leather hose, the left side of the outer wall of the connecting leather hose is fixedly connected with a fixed shell, the inner wall of the fixed shell is fixedly connected with a filter screen, and the right side of the outer wall of the filter screen is fixedly connected with a sealing gasket. A butt joint pipe is fixedly connected to the left side of the outer wall of the sealing gasket, and a fixing mechanism is fixedly connected to the right side of the outer wall of the connecting leather pipe and used for preventing gas and water vapor from flowing back. The leather hose is connected to the right side of the pneumoperitoneum machine, the fixing shell with the filter screen is fixed to the left side of the leather hose, a moisture filter is added to keep dry, a butt joint pipe on the left side of the sealing gasket is convenient to replace, the pipe wall of the connecting cylinder is extended, the disposable leather hose is convenient to replace, safety is guaranteed, the puncture outfit is rapidly fixed through the leather hose connector, and the risks of blurring and hydrops of a laparoscope are reduced; gas quality is ensured, infection is prevented, and postoperative health of patients is guaranteed.