534results about How to "Improve the correction effect" patented technology

The invention discloses a construction site digital-to-analog conversion and collaborative management method based on a distributed computing architecture. The method comprises the following steps: S1, collecting microscopic level data of a construction site; s2, obtaining mid-level data of the construction site; s3, acquiring macroscopic hierarchical data, and constructing a multi-scale spatio-temporal data set; s4, decomposing and processing the multi-scale spatio-temporal data set by using a spatio-temporal diagram convolutional network, and constructing a dynamic digital twin model; s5, designing a competition and cooperation mechanism between agents by using a game theory model, solving resource allocation conflicts through Nash equilibrium, and dynamically optimizing resource scheduling; s6, optimizing a scheduling strategy of the intelligent agent through a distributed algorithm based on near-end strategy optimization; s7, detecting errors in digital-to-analog conversion and scheduling processes by using a belief propagation algorithm; and S8, optimizing a scheduling strategy through a distributed adaptive feedback control mechanism. According to the invention, a distributed computing architecture, space-time diagram convolution, an intelligent scheduling technology and the like are adopted, and digital-to-analog conversion and resource management of the construction site are optimized.

The invention discloses a display screen image visual angle unevenness correction method and device and a storage medium, which are used for improving the optical measurement precision of a display screen Mura inspection and compensation system. Aligning the central position of the sampling camera to the central position of the calibration screen; inputting the visual angle non-uniform calibration picture with the gray scale of g into a calibration screen; detecting an ROI (Region of Interest) on the calibration screen to generate a brightness array; inputting the pure-color picture with the gray scale of g into the calibration screen, and acquiring the pure-color picture through the sampling camera to generate a pure-color display screen image; calculating a gray average value of the pure color display screen image in the ROI region, and generating a gray average value array; performing centering normalization processing on the brightness array and the gray average array to generate normalized data; calculating a correction coefficient set according to the normalized brightness mean value data and the normalized gray mean value data; generating a group of distance arrays; and according to the distance array and the correction coefficient set, constructing a visual angle non-uniform model of the calibration screen under the gray scale g.

The invention belongs to the technical field of steel cofferdam displacement monitoring, and particularly discloses a deepwater bottom-sealing-free double-wall steel cofferdam construction displacement monitoring system which is characterized in that multi-source displacement monitoring is carried out on a lowered cofferdam section in a test monitoring time period, water level and flow speed monitoring is synchronously carried out, an environment influence monitoring source is identified by means of displacement mutability analysis, and the construction displacement of a deepwater bottom-sealing-free double-wall steel cofferdam is monitored. Therefore, the effective monitoring source is selected to perform accurate displacement monitoring in combination with the monitoring results of the water level and the flow velocity in the subsequent displacement monitoring, the process realizes the effect verification before the implementation of the displacement monitoring, the accuracy of the displacement data is ensured to the greatest extent, and meanwhile, when the displacement monitoring result is used for analysis, the accuracy of the displacement data is improved by identifying the offset monitoring point. The arrangement positions and the deviation directions of the monitoring points are combined, deviation guiding analysis of the lowered steel cofferdam section is carried out, deep utilization of displacement monitoring data is achieved in the process, targeted guidance can be provided for displacement deviation correction, and the deviation correction efficiency and the deviation correction effect can be improved easily.

The invention provides an auxiliary graph correction method and device, a medium and a product, and the method comprises the steps: determining at least one target edge in an initial layout according to the graph attribute of each graph in the initial layout; a recognition area is constructed with the first edge as the benchmark, and the first edge is any one of the at least one target edge; under the condition that the main graph exists in the recognition area, adding an initial auxiliary graph between the main graph and the first edge; judging whether the initial auxiliary pattern conforms to a preset mask manufacturing rule or not; and if not, correcting the initial auxiliary pattern according to a preset mask manufacturing rule to obtain a target auxiliary pattern. The embodiment of the invention can improve the correction effect of the auxiliary graph.

The utility model relates to the technical field of composite film preparation, and discloses a composite film rewinding device which comprises a bottom plate, two sets of supporting plates are symmetrically installed on the top of the bottom plate, winding drums and tensioning rollers are arranged on the supporting plates, a supporting beam is fixedly installed between the supporting plates, a telescopic air cylinder is installed on the supporting beam in a penetrating mode, and the telescopic air cylinder is fixedly installed on the bottom plate. A supporting frame is fixedly installed at the bottom of the telescopic air cylinder. According to the deviation rectifying device, the first forward and reverse screw rod is rotated to be matched with the action of the guide block and the guide groove, the first forward and reverse screw rod drives the two first deviation rectifying plates to move reversely so as to adjust the distance between the two first deviation rectifying plates, the second forward and reverse screw rod is rotated to drive the two second deviation rectifying plates to move reversely, and therefore the deviation rectifying effect on the composite film can be achieved; and the first deviation rectifying plate and the second deviation rectifying plate are limited through the first positive and negative screw rod and the second positive and negative screw rod correspondingly, deviation caused by vibration is prevented, and the deviation rectifying effect is optimized.

The invention relates to an image correction technology, and discloses a CBCT (cone beam computed tomography) image osteosclerosis artifact correction method, which comprises the following steps: acquiring a high-energy projection image and a low-energy projection image of a target object, performing image reconstruction, and calculating to obtain a primary artifact-free image; extracting a bone image and a soft tissue image from the primary artifact-free image based on threshold segmentation, calculating a bone weight, taking the artifact image of the target object as training input data, taking the bone image and the soft tissue image as reference data, and training to obtain a dual-channel artifact correction model; and outputting a preprocessed bone image and a preprocessed soft tissue image according to the to-be-processed osteosclerosis artifact image by using a dual-channel artifact correction model, and carrying out weighted fusion on the preprocessed bone image and the preprocessed soft tissue image according to a bone weight to obtain an artifact-free reconstructed image. The invention further provides a CBCT image osteosclerosis artifact correction device, electronic equipment and a storage medium. According to the invention, the image artifact correction effect can be improved.

The invention discloses a map moving track deviation correction method and system, belongs to the technical field of electronic maps, and aims to solve the technical problems that in the prior art, deviation exists in GPS track data, so that position information in map application is not accurate, and the map moving track deviation is not accurate. According to the technical scheme, the method comprises the following steps: data acquisition: receiving original latitude and longitude coordinate data and sensor data from a GPS satellite; data preprocessing: denoising and missing value processing are performed on the original GPS trajectory data, and the data quality is improved by using a Kalman filtering algorithm; feature extraction: extracting spatio-temporal features from the preprocessed data; constructing a deep learning model: constructing a deep neural network model for learning the spatial pattern and time sequence features of the trajectory data; the deep learning model inputs the spatial pattern and the time sequence characteristics of the trajectory data, and outputs the trajectory point coordinates after deviation correction; training and optimizing; correcting deviation in real time; and dynamically adapting.

The utility model relates to the technical field of belt conveyors, and discloses a powerful automatic deviation rectifying device for a belt conveyor, which comprises a base, a roller bracket rotationally arranged in the middle of the top end of the base through a supporting column, a deviation rectifying component arranged on one side of the supporting column, a middle roller arranged on the top of the roller bracket, and fixing seats arranged on two sides of the top end of the roller bracket. A mounting seat is rotationally arranged in the fixed seat, a side roller is rotationally arranged in the mounting seat, and an adjusting assembly is arranged in the mounting seat; by means of the arranged adjusting assembly, the angles of the mounting base and the side rollers can be adjusted adaptively, conveying belts of different specifications and sizes can be used conveniently according to needs, the use range is widened, the deviation rectifying effect of the conveying belts is improved, and by means of the arranged deviation rectifying assembly, the supporting columns can rotate, so that the deviation rectifying efficiency is improved. And meanwhile, the rotating roller bracket is driven to rotate and adjust, the middle roller and the side roller at the top of the rotating roller bracket are matched, powerful automatic deviation rectifying operation is conveniently conducted on a conveying belt of the conveyor, and the using effect is good.

The invention relates to the technical field of image processing, and provides a high-precision image splicing and correcting method based on a multi-view camera, and the method comprises the steps: recognizing a potential distortion region in an image according to a plurality of view images, and outputting the distortion region; detail features of the distortion areas are extracted, the distortion type of each distortion area is derived by combining geometric consistency of adjacent areas, and local correction parameters are generated; performing accurate correction on each pixel by adopting self-adaptive nonlinear transformation and combining a mapping relation between a pixel position and a distortion degree, and obtaining an image after distortion correction; and after image splicing is carried out on the images after distortion correction, the transition effect of the edge is optimized by combining a multi-scale region fusion technology. According to the invention, the visual unnatural phenomenon at the seam can be eliminated, and the precision and visual experience of the whole image splicing are improved.

Compared with the prior art, the profile steel straightening machine disclosed by the invention has the advantages that the bending degree detection unit is arranged to carry out bending degree detection on a web plate which is straightened by the web plate straightening arm, and the heating unit is used for heating a part which is detected by the bending degree detection unit and has relatively large bending degree; then the heated bent position is extruded and corrected through the correction unit, internal stress can be released through heating, and therefore the bent position can be corrected more easily, and a better correction effect is achieved; comprising a web correction device and a plurality of groups of web correction arms arranged at intervals on the web correction device, and further comprises a curvature detection unit arranged between two adjacent groups of web correction arms and used for detecting the curvature of a web; the heating unit is arranged on the downstream of the curvature detection unit and used for heating the area, exceeding the preset value, of the curvature of the web; and the correcting unit is used for correcting the heated area of the web plate.

The invention discloses an ophthalmology department vision examination data collection and analysis system, and relates to the technical field of ophthalmology department vision examination data collection and analysis. An examination data collection module is used for examining and collecting ametropia and eye movement behaviors of a current patient, and an ametropia basic correction unit is used for calculating and outputting correction lens degree JD; an eye movement capturing precision evaluation value Y is calculated and output by utilizing an eye movement capturing precision evaluation unit, a feedback adjustment coefficient T and an updated correction lens degree JDnew are sequentially calculated and output by utilizing an ametropia feedback adjustment correction unit, and a result display module is utilized to display the correction lens degree JDnew. The updated correction lens degree JDnew of current correction and next optometry is displayed to the current patient, correction degree calculation of ametropia is continuously optimized according to the accuracy change of eye movement behavior capture, then the correction effect is improved, meanwhile, a formed feedback cycle mechanism also enables the system to have the self-learning and optimization ability, and the system can be widely applied to the field of medical science, medical science and medical science. And individual differences and disease changes of patients can be continuously adapted.

The invention relates to the technical field of power grid fault positioning, in particular to a fault detection optimization method for a honeycomb-shaped smart power grid. The method comprises the following steps: acquiring an original data sequence in a smart power grid, decomposing the original data sequence by using a time sequence decomposition algorithm to obtain an initial residual component sequence, and determining the residual component sequence according to the data distribution characteristics in the initial residual component sequence and the data difference between the initial residual component sequence and the original data sequence. Obtaining the credibility of the initial residual component sequence; if the credibility of the initial residual component sequence is not within a preset credibility range, optimizing preset parameters in a time sequence decomposition algorithm until the credibility of the new residual component sequence is within the preset credibility range, and obtaining a target residual component sequence; and finally, performing fault detection on the smart grid according to the target residual component sequence. By optimizing the preset parameters in the time sequence decomposition algorithm, the fault detection accuracy of the residual component is improved.

The invention provides a document image correction method based on deep learning, and belongs to the field of computer vision, and the method comprises the steps: carrying out texture rendering and offset mapping processing based on an original texture grid image in a training set, and generating a spatial image to be rendered; performing data enhancement on the spatial to-be-rendered image through a flat image corresponding to the original texture grid image, and performing random disturbance based on a transformation space to obtain a corresponding sample training pair; based on the sample training pairs in the training set, training the initial detection correction model, optimizing control parameters, and obtaining a detection correction model; detecting a four-point detection model and a document correction model in the correction model, and processing the document image to be corrected in sequence to generate a corrected output image; according to the method, model training is carried out based on the self-constructed sample training pair, the document image correction effect of the model is effectively improved, meanwhile, the method can adapt to a complex real scene, and subsequent detection and recognition of the document image are facilitated.

The invention discloses a color constancy method based on wavelet attention and a VMama network, and belongs to the field of computer vision and image enhancement. According to the method, illumination parameter estimation is carried out by adopting a designed WA-VMama network, and then color correction is carried out by adopting the estimated illumination parameters, so that a corrected image under standard illumination is obtained. According to the method, a designed wavelet attention mechanism is embedded into VMama, and a deep network model (WA-VMama for short) is constructed, so that global information can be effectively extracted by using the network, local information can be considered, the method has higher feature extraction and expression ability, and the accuracy of illumination parameter estimation is improved.

The invention relates to a tower drum hoisting positioning device which comprises a first deviation rectifying assembly, a second deviation rectifying assembly and a third deviation rectifying assembly. The first deviation rectifying assembly is suitable for being detachably installed at the top of the lower barrel section and the bottom of the upper barrel section and used for rectifying the falling direction of the upper barrel section. The second deviation rectifying assembly is suitable for being detachably mounted at the top of the lower barrel section and used for rectifying the falling direction of the upper barrel section; the third deviation rectifying assembly is suitable for being detachably installed at the top of the lower barrel section and the bottom of the upper barrel section and used for rectifying the falling direction of the upper barrel section. On the whole, manual work can be assisted or manual work can be replaced for correcting the falling direction of the upper barrel section, the correcting effect is good, potential safety hazards are greatly reduced, and the labor intensity and difficulty are reduced. Moreover, the second deviation rectifying assembly and the third deviation rectifying assembly are detachably connected with the drum section, and after the lower drum section and the upper drum section are assembled, the second deviation rectifying assembly and the third deviation rectifying assembly can be detached, so that the second deviation rectifying assembly and the third deviation rectifying assembly can be recycled in the whole tower drum assembling process, and the assembling cost is reduced.

The invention discloses an image distortion correction method, device and system and a storage medium, which are used for improving the aliasing elimination effect of a display screen image. The method comprises the following steps: partitioning a display screen image to be corrected to generate a plurality of image sub-block sets; inputting the plurality of image sub-blocks into a scale-invariant distortion correction model for distortion correction, and generating a plurality of transformation matrixes which are transformation coefficient sets of pixel point coordinates before and after correction; constructing an adaptive convolution filter according to the plurality of transformation matrixes; inputting the plurality of image sub-blocks after distortion correction into the adaptive convolution filter for anti-aliasing convolution and interpolation calculation, and outputting the plurality of image sub-blocks after pixel point optimization correction; performing multi-scale anti-mixing correction on the edge areas of the plurality of corrected image sub-blocks; and splicing the plurality of image sub-blocks to generate a complete target correction image.

The invention relates to an intelligent torque correction method for a hydraulic motor, and the method comprises the steps: obtaining the torque generated by a liquid inlet plunger in a hydraulic part, and calculating the real-time hydraulic output torque of the hydraulic part; an actual torque deviation value is calculated based on the real-time motor output torque and the real-time hydraulic output torque, an actual torque loss rate is calculated through an efficiency loss model based on the actual torque deviation value and the total input power of the hydraulic motor, the lubricating liquid film thickness in the hydraulic part is obtained, and a loss-liquid film data set is constructed; the liquid film deviation factors are input into a preset deviation analysis model, and the deviation analysis model carries out analysis to generate the liquid film deviation factors; and inputting the historical actual torque loss rate and the historical liquid film deviation factor into a preset torque deviation analysis model to analyze and generate a predicted torque loss rate, and calculating a predicted torque deviation value based on the motor output torque and the predicted torque loss rate. The method has the effect of improving the prediction accuracy of the torque of the hydraulic motor.

The invention provides a composite phase-locked loop for a linear frequency sweep laser. The composite phase-locked loop comprises a frequency modulation continuous wave source, a Mach-Zehnder interferometer module, a balance detector and an optical phase-locked loop module. Wherein the frequency modulation continuous wave source is used for generating frequency modulation continuous waves; the Mach-Zehnder interferometer module is arranged on a laser light path and receives a laser beam through an optical fiber patch cord; the balance detector is arranged on a laser light path and is used for receiving the laser beam, converting the laser beam into a beat frequency electric signal and inputting the beat frequency electric signal into the optical phase-locked loop module; the optical phase-locked loop module is electrically connected to the frequency modulation continuous wave source and the Mach-Zehnder interferometer module, and the optical phase-locked loop module, the frequency modulation continuous wave source and the Mach-Zehnder interferometer module form a feedback loop. According to the composite phase-locked loop architecture, the performance of a single phase-locked loop can be further improved, and due to the unique connection mode of the auxiliary loop, the scheme can eliminate the influence of environmental noise or vibration on the system stability to a certain extent.

The invention relates to the technical field of intelligent sensing thunder and lightning interceptors, in particular to an intelligent thunder and lightning interceptor and a lightning surge pulse protection method, and the method comprises the steps: measuring a thunder and lightning current sequence when thunder and lightning occur through a thunder and lightning current recorder, dividing the thunder and lightning current sequence to obtain thunder and lightning current subsequences, and calculating the thunder and lightning change rate of each thunder and lightning current; determining the lightning magnetic field deviation degree of each lightning current based on the difference between the lightning change rates of each lightning current and the adjacent lightning current and the average value of the lightning change rates of all lightning currents in the lightning current sub-sequence, and obtaining the lightning magnetic field interference intensity of each lightning current by combining the difference between each lightning current and the average value of the left and right adjacent lightning currents. Calculating a lightning magnetic field correction coefficient of each lightning current, and correcting each lightning current in the lightning current sub-sequence; and intercepting thunder and lightning by adopting a thunder and lightning interceptor according to the corrected amplitude spectrum and energy spectrum of the thunder and lightning wave. The lightning current data precision can be improved, and the lightning interception accuracy is ensured.

The invention discloses a thermal forming correction control system of a non-standard mineral casting, and belongs to the field of intelligent control, and the thermal forming correction control system comprises the following steps: obtaining design requirements of the non-standard mineral casting to obtain a casting segmentation result; according to material requirements, a sample data set is constructed through multi-dimensional experimental data, and learning is carried out to obtain a parameter control analysis model; multi-dimensional parameter analysis is carried out, and control parameters of each casting segmentation area are determined; based on the control parameters, acquiring an abnormal case set, performing abnormal probability calculation, converting the abnormal case set into a control coefficient, performing abnormal parameter tracing, and determining abnormal control parameters; and thermal forming correction control information is obtained and sent to the controller for control. According to the non-standard mineral casting thermal forming correction control method, the technical problem that the correction effect is poor due to the fact that uniform parameters are adopted for existing non-standard mineral casting thermal forming correction control is solved, and the technical effects that the non-standard casting is subjected to differentiated segmentation, different areas are correspondingly controlled and the overall correction effect is improved according to geometrical characteristics and material characteristics are achieved.

The invention relates to the technical field of image processing, in particular to a global-to-local answer sheet image intelligent correction method and device. The method comprises the following steps: correcting a to-be-corrected answer sheet image to obtain a target answer sheet image based on first mark point position information with the highest matching degree with a mark point image in an answer sheet template image and corresponding second mark point position information in the to-be-corrected answer sheet image; determining a target straight line corresponding to each edge of the question template frame in the target answer sheet image as an edge of a question target frame; matching a title target frame corresponding to the title template frame in the target answer sheet image; determining a moving distance corresponding to the target answer sheet image based on a first offset of the target straight line relative to the corresponding edge and / or a second offset of the title target frame relative to the title template frame; and moving the black pixel points in the question target frame and / or the title target frame according to the moving distance. According to the invention, the deformation problem of the answer sheet image can be corrected from global to local, and the correction effect is improved.

The present disclosure relates to an image correction method and device, an electronic device and a storage medium. The image correction method comprises: obtaining a first image captured by an image acquisition module; obtaining a mapping diagram generated based on the corresponding relationship between the field of view and the distortion rate of the image acquisition module; wherein the mapping diagram comprises: a mapping relationship between the original pixel coordinates of each pixel value in the original image captured by the image acquisition module and the corrected target pixel coordinates; based on the mapping diagram, correcting the first image to obtain a corrected second image.

The invention provides an unmanned aerial vehicle image real-time correction method, system and device and a storage medium, and relates to the technical field of image correction, and the method comprises the steps: obtaining the camera pose information, image information and shooting space information of an unmanned aerial vehicle image; performing image distortion correction on the unmanned aerial vehicle image based on the camera pose information and the image information of the unmanned aerial vehicle image to obtain a distortion-corrected unmanned aerial vehicle image; and performing geometric correction on the unmanned aerial vehicle image subjected to distortion correction based on the camera pose information, the image information and the camera shooting space information of the unmanned aerial vehicle image so as to obtain an unmanned aerial vehicle image subjected to geometric correction. According to the invention, image distortion correction and geometric correction are carried out on the unmanned aerial vehicle image according to the camera pose information, the image information and the photography space information of the unmanned aerial vehicle image, so that the problem of large image correction error caused by a non-system due to uncontrollable non-system factors in real-time image correction of an existing unmanned aerial vehicle is solved.

The utility model discloses a scoliosis orthosis, and relates to the technical field of spine orthosis instruments. Comprising two arc-shaped waist plates, the arc-shaped waist plates are symmetrically arranged left and right, moving sleeves sleeve the arc-shaped waist plates, a first mounting mechanism is arranged on a left vertical plate, second fixing mechanisms are arranged on the right side face of the first mounting mechanism close to the front and back positions, and a first correcting arc-shaped plate is mounted between the right sides of the front and back second fixing mechanisms; a first prompting mechanism is arranged on the right side face of the first correcting arc-shaped plate. The distance between the arc-shaped waist plates can be adjusted, the height of the first arc-shaped correcting plate and the height of the second arc-shaped correcting plate can be adjusted, so that the use of various people can be met, the practicability is high, the correcting force can be displayed, the correcting force can be adjusted gradually, the correcting effect is better, and the practicability is high. And the problem that the correction force is too large or too small is avoided, and use is facilitated.

The invention discloses a special vehicle framework intelligent shape righting system and method based on dynamic measurement control, and the system comprises a movable support, a ground scanning module disposed on a front side supporting leg of the movable support, and a welding line recognition module disposed at the center of the front side of the top of the movable support. And the deformation correction mechanism is arranged at the top of the movable bracket and can move left and right. According to the invention, different distance values of different positions are dynamically obtained by controlling a plurality of groups of distance measurement identification modules, welding seam positions are distinguished and identified, to-be-corrected positions are determined, and the height of the supporting legs is controlled to rise and fall in combination with the ground variation amplitude, so that automatic and intelligent identification control and judgment can be carried out on the special vehicle framework at any place; therefore, intelligent shape righting control of the whole special vehicle framework is realized.

The embodiment of the invention provides a method and device for measuring the thickness of a wafer metal film in the polishing process, polishing equipment and a storage medium. The method for measuring the thickness of the wafer metal film in the polishing process comprises the following steps: acquiring a current temperature and a current measurement signal; wherein the current temperature is used for indicating the temperature of a metal film of the wafer at the current moment; the current measurement signal is collected by an eddy current sensor, and the current measurement signal is related to the thickness of the metal film; determining the measurement thickness of the metal film according to the current measurement signal; correcting the measured thickness according to the current temperature and the temperature correction information to obtain the thickness of the metal film; the temperature correction information is used for indicating the relation among the current temperature, the measured thickness and the thickness of the metal film.

The invention relates to the technical field of winding rollers, in particular to a decorative cloth production winding device with a deviation rectifying mechanism, which comprises a winding roller used for winding decorative cloth, deviation rectifying baffles sleeve two ends of the winding roller, air collecting cavities are formed in the deviation rectifying baffles, a plurality of air dispersing holes are formed in the surfaces of the deviation rectifying baffles, and the air dispersing holes are communicated with the air collecting cavities. A containing groove is formed in the outer ring face of the deviation rectifying baffle, a micro air pump is fixed into the containing groove, an air conveying pipe is installed at the air outlet end of the micro air pump and extends into the air collecting cavity, and a cleaning brush plate is installed on one side of the deviation rectifying baffle. The winding device has the beneficial effects that the micro air pump and the air collecting cavity are arranged in the deviation rectifying baffle, high-pressure airflow is dispersed through the air dispersing holes, soft flocks and dust on the surface of the winding roller are effectively blown off, the winding roller is kept clean, dust in the air is effectively blocked through the design of the dustproof cloth belt, and the service life of the winding device is prolonged. And the blown airflow is clean and pollution-free.

The utility model provides a circular knife die-cutting machine with a buffering function, and relates to the technical field of circular knife die-cutting machines, the circular knife die-cutting machine comprises a feeding platform, a material belt buffering device, an adjustable deviation rectifying corona device, a die-cutting device and a conveying device which are arranged in sequence, the material belt buffering device comprises a fixed roller set and a sliding roller set, the fixed roller set comprises a plurality of fixed guide rollers, and the sliding roller set comprises a plurality of sliding guide rollers. The sliding roller set comprises a lifting seat, a lifting driving mechanism and a plurality of sliding guide rollers, a material belt bypasses each fixed guide roller and each sliding guide roller at intervals, the adjusting type deviation rectifying corona device comprises a deviation rectifying assembly and a corona assembly, the deviation rectifying assembly is connected with a sliding adjusting mechanism, the corona assembly is connected with a drawing adjusting mechanism, and the drawing adjusting mechanism is connected with the lifting seat. According to the circular knife die-cutting machine, non-stop material changing of the circular knife die-cutting machine is achieved through the arrangement of the material belt temporary storage device, meanwhile, the deviation rectifying effect on the material belt is improved, and the yield of products is guaranteed.