74results about How to "Retain structure" patented technology

The invention discloses an image denoising method based on double-branch feature fusion, and the method specifically comprises the steps: 1, carrying out the collection of an image, and carrying out the preprocessing of the collected image; 2, a denoising network with double-branch feature fusion is built, the network comprises a generation network and a discrimination network, and authenticity discrimination is carried out on a denoised image generated by the generation network and a real image through the discrimination network; step 3, adversarial training is carried out on the generative network and the discriminant network to obtain a trained dual-branch feature fusion denoising network; and step 4, performing denoising processing on the image by using the trained generative network. The method is remarkably superior to a traditional denoising method in denoising quality, model generalization ability and operation time, different types of noise can be effectively removed, and more image details and structure information are reserved.

The invention belongs to the technical field of remote sensing monitoring, and particularly relates to a sequential SAR crop classification method based on lightweight linear attention. The method comprises the following steps of: firstly, efficiently extracting spatial characteristics of an image of each time step in a time sequence image by utilizing a weight-shared U-Net encoder by utilizing spatial and temporal characteristics of the time sequence SAR image; lLTAEBlock is used for extracting spatiotemporal features from time sequence advanced spatial features, FAVOR + is used for achieving linear calculation complexity, a shared query matrix serves as a learnable parameter, the calculation amount and the parameter amount are reduced, global information interaction in the time dimension is achieved, and modeling is conducted on the complex phenological relation of crops; and the decoder retains the deep spatial-temporal features and fuses the time sequence shallow spatial features to finally obtain a prediction result with the same spatial scale as the input data. And under the constraint of the loss function, adjusting through an optimizer to obtain a crop classification model with the best fitting. The method is especially suitable for crop classification of long-time-sequence SAR images, reduces the need of a memory, and improves the crop classification efficiency.

The invention discloses an advertisement compliance knowledge question-answering method, device and equipment for foreign enterprises, and the method comprises the steps: obtaining advertisement compliance data of a global data source, carrying out the document structuring and knowledge extraction of the advertisement compliance data, and constructing an advertisement compliance knowledge graph; converting text data in the advertisement compliance knowledge graph into vectors to obtain vector data; semantic analysis is conducted on a query question input by a user, construction is conducted in combination with the vector data, prompt words are obtained, the prompt words are input into a preset large language model, and target answers are obtained; and outputting the target answer based on the advertisement compliance knowledge graph, and visually displaying the target answer in a corresponding answer query graph. And comprehensive and efficient compliance information query and answer services can be provided, so that compliance review requirements of foreign enterprises on advertisements in a multi-language and cross-regional environment can be met.

The invention discloses a feature weighted fusion hyperspectral image classification method based on GAT and CNN. The method comprises the following steps: firstly, dividing a hyperspectral data set into a training set, a test set and a verification set; then, a double-branch feature weighted fusion network model combining CNN and GAT is constructed; inputting the divided training set into a classification network model, setting parameters, and training to obtain training indexes; and finally, evaluating indexes, data and image information of a classification result. According to the method, the Euclidean and non-Euclidean feature information of the hyperspectral image is fully extracted by utilizing the capture capability of the depth separable convolution and the GAT on the pixel information and the super-pixel data, and the accuracy and generalization capability of classified data are improved.

The invention discloses a double-criterion abnormal electricity consumption detection method based on dimension reduction clustering, and relates to the technical field of electricity consumption detection, and the method comprises the following steps: building a daily and monthly electricity consumption data set of electricity consumption users based on original intelligent electric meter data, constructing feature indexes, and carrying out the normalization to obtain a high-dimensional daily and monthly electricity consumption feature set; performing dimensionality reduction processing on the high-dimensional daily and monthly electricity consumption feature set by using a uniform manifold approximation and projection algorithm to obtain a low-dimensional projection feature set, and setting a double-criterion detection rule based on an unsupervised clustering algorithm to construct an abnormal electricity consumption detection model; and based on the abnormal power consumption detection model, inputting the low-dimensional projection feature set, outputting an abnormal power consumption detection result, and evaluating output by using a model evaluation index. According to the method, data can be better clustered, the parameter selection difficulty is reduced, and accurate optimal clustering center selection is realized, so that the accuracy of an abnormal power utilization detection model is improved.

The invention discloses a large-scale MIMO (Multiple Input Multiple Output) CSI (Channel State Information) feedback method fused with a self-attention mechanism, which is fused with the self-attention mechanism and is combined with residual connection and a deep separable convolutional layer to realize efficient and accurate CSI feedback. In the encoder part, a multi-layer self-attention mechanism is adopted to carry out feature extraction and compression; in a decoder part, through cooperative work of multiple convolutional layers and a self-attention mechanism, an embedded vector generated by an encoder is reconstructed into an original image. The model performance is remarkably enhanced by combining residual connection and a depth separable convolutional layer sequence in an attention mechanism. In the encoder part, a multi-layer self-attention mechanism is adopted to realize efficient feature extraction and compression. In a decoder part, multiple convolutional layers and a self-attention mechanism work cooperatively, and an embedded vector generated by an encoder is reconstructed into an original image. The method can be used for efficiently and accurately realizing the CSI feedback task of the large-scale MIMO system.

The invention belongs to the technical field of plant drinks, and particularly relates to kidney-tonifying and bowel-relaxing cistanche tea and a preparation method thereof. The herba cistanche tea is prepared from the following raw materials in parts by weight: 30 to 40 parts of herba cistanche, 10 to 15 parts of fructus lycii extract, 10 to 12 parts of semen cassiae extract, 10 to 15 parts of fructus cannabis extract, 8 to 12 parts of radix polygoni multiflori extract, 8 to 10 parts of rhizoma dioscoreae extract, 3 to 5 parts of honey powder, 5 to 10 parts of maltodextrin, 2 to 3 parts of beta-cyclodextrin, 2 to 3 parts of antioxidant, 0.1 to 0.15 part of mixed enzyme and 2 to 3 parts of zymophyte. According to the preparation method disclosed by the invention, the flowers and the rhizomes of the cistanche are subjected to pretreatment and enzymolysis treatment in different modes, and then fermentation is performed under the same microbial fermentation condition, so that respective advantages of the two parts of raw materials are fully played, beneficial components are released and retained to the maximum extent, and the functionality and the quality of the product are remarkably improved through synergistic fermentation.

The invention discloses a preparation method of a pre-fibration polymer solid electrolyte as well as a product and application of the pre-fibration polymer solid electrolyte. The method comprises the following steps: demulsifying and drying polytetrafluoroethylene (PTFE) emulsion to prepare pre-fiberized PTFE dry powder, then mixing sulfide lithium phosphorus sulfur chlorine (Li6PS5Cl) electrolyte and fiberized PTFE in a proper proportion, and directly preparing the sulfide solid electrolyte composite membrane through a room temperature rolling solvent-free dry process. The pre-fiberized solid sulfide electrolyte film avoids high-cost and high-energy-consumption operations such as ball milling, hot pressing and the like, and the preparation energy consumption and the equipment cost are reduced. The material can be effectively applied to the field of preparation of solid-state batteries, and the structure and performance of an electrolyte material are fully reserved. The method is a solid electrolyte film forming technology with an application prospect, and has an important application value for the development of a solid-state battery.

The invention discloses a fault diagnosis method based on time sequence characteristic linear modulation intelligent oversampling, and the method supplements the details of a high frequency band through the improvement of the signal sampling rate, and improves the fault diagnosis performance. The method comprises the steps of obtaining an original vibration signal; an original vibration signal is preprocessed, and a low-sampling-rate signal is obtained through down-sampling. And pairing the low-sampling-rate signal and the original vibration signal for intelligent super-sampling network training. And a super-sampling rate model is constructed, and prediction output of super-sampling rate signals is realized. And constructing a fault diagnosis model, and learning features of signal fault classification. And inputting the training set into a super-sampling rate fault diagnosis model to train the model. And inputting a vibration signal into the trained model to obtain an oversampling rate signal, and classifying the oversampling rate signal to obtain a health state type of the oversampling rate signal. The fault diagnosis model provided by the invention better solves the problem that the diagnosis performance is poor when facing low-sampling-rate signals, and meanwhile, the high-frequency information of the signals is relatively truly recovered.

The invention discloses a maleamic acid-tragacanth copolymer as well as a preparation method and application thereof, and the maleamic acid-tragacanth copolymer is prepared from the following raw materials in parts by weight: 3-7 parts of maleic anhydride, 90-110 parts of an absolute ethyl alcohol solvent, 10-15 parts of polyethyleneimine, 8-10 parts of polyethylene polyamine and 40-60 parts of an alcohol-water solution, and 15-30 parts by weight of a tragacanth modified derivative. A tragacanth modified derivative is introduced into maleic anhydride amidation in a ring-opening grafting blending manner, so that two adsorption materials have a synergistic effect, the characteristics of biodegradability, a macromolecular porous structure and the like are retained, and a large number of functional groups such as hydroxyl, carboxyl, amido, carbon-carbon double bonds and the like are introduced; toxic substances such as heavy metals, medical intermediates, dyes and pesticides in the wastewater are removed by virtue of comprehensive effects of electrostatic force, hydrogen bonds, Van der Waals force, inclusion and the like.

The invention discloses a scene-based push-broom hyperspectral image adaptive non-uniformity correction method, which comprises the following steps of: automatically sorting each scanning line of a wave band of a hyperspectral image from small to large according to pixel DN values of different spatial dimensions, and adaptively selecting at least one dark uniform region and one bright uniform region of the spatial dimensions; a correction coefficient of each spatial dimension pixel is obtained, and non-uniform correction of each spatial dimension is completed; on the basis of spectral dimension characteristics, spectral dimension non-uniform correction is completed according to the spatial dimension non-uniform correction result, spatial dimension non-uniform correction is repeated according to the spectral dimension non-uniform correction result, and hyperspectral image non-uniform correction is completed. The method is applied in a full-automatic mode, limitation of scenes can be broken through, space and spectrum combined information is used for non-uniform correction, space structure information and spectrum information of the hyperspectral images are reserved to the maximum extent, the quality of the hyperspectral images can be effectively improved, and the method is suitable for self-adaptive non-uniform correction of mass hyperspectral images of various complex scenes. And the method has good universality.

The invention relates to the technical field of semantic segmentation, in particular to a night semantic segmentation method based on low illumination enhancement and edge optimization, and the method comprises the steps: inputting an image into a low-light enhancement repair network based on the Retinex theory, local contrast enhancement and adaptive feature fusion, obtaining a denoised and enhanced intermediate image, and carrying out the edge optimization of the intermediate image; inputting the intermediate image into a semantic segmentation network to obtain a category distribution diagram of each pixel; inputting a discriminator embedded with a channel attention module according to the category distribution diagram of each pixel, and optimizing a generator composed of a low light enhancement repair network and a semantic segmentation network through a multi-task joint optimization loss function; and inputting a night image to be detected and segmented into the optimized generator, and outputting a segmentation result. By adopting the method, the low-light enhancement repair network is combined with a local contrast enhancement and channel feature fusion mechanism, so that the overall brightness of the image is improved, the details and edge structures of the image are reserved, and the perception capability and robustness of the model are improved.

The invention discloses a polarization degree and light intensity image fusion method based on bidirectional cross attention. The method belongs to the technical field of polarization imaging, computational imaging and image processing. The technical problems that in the prior art, dynamic balance between structural detail keeping and polarization saliency enhancing is generally difficult to achieve, and particularly under the condition of a complex background or a low signal-to-noise ratio, texture missing or noise artifacts are prone to occurring in a fusion result are solved. A fusion mechanism capable of establishing a two-way information interaction relationship between a light intensity image and a polarization degree image is provided, and further processing is carried out aiming at a noise problem, so that more stable and more reliable polarization fusion imaging is realized.

The invention relates to the technical field of logistics management, in particular to an in-transit supervision method for large-size transportation on an expressway. The method comprises the following steps: acquiring transportation speed data and transportation vibration data; carrying out transportation speed feature extraction on the transportation speed data to obtain transportation speed feature data, and carrying out transportation vibration feature extraction on the transportation vibration data to obtain transportation vibration feature data; performing feature map construction on the transportation speed feature data and the transportation vibration feature data to obtain transportation speed vibration feature map data; abrupt change feature collection and change degree extraction are carried out according to the transportation speed vibration feature map data, and transportation abrupt change feature data and transportation change degree data are obtained so as to carry out large piece transportation anomaly detection operation. According to the invention, the safety and the management level of the large-piece transportation process are effectively improved.

The invention discloses a method for protecting a privacy account based on big data joint training, and relates to a method for protecting the privacy account. The invention aims to solve the problem that a traditional privacy account protection means cannot really achieve the purpose of protecting a privacy account of a user. The method comprises the following steps: establishing a user account data management system; inputting user common login position information and password input speed information into a user account data management system, and training the input information; when the user logs in the account, the user account data management system compares the common login position information of the user with the password input speed information, and after the comparison succeeds, a verification image is displayed; performing sharpness processing on the verification image; after verification succeeds, the user is allowed to enter, otherwise, the abnormal condition transmitting module transmits abnormal condition information to the user; and after the user confirms the abnormal condition information, authorization information is transmitted to the abnormal condition user confirmation module. The invention belongs to the technical field of network account security.

The invention discloses a meal replacement composition for regulating blood sugar balance and controlling body weight and a preparation method thereof. The meal replacement composition is comprehensive in nutrition, can achieve a good satiety effect, effectively delays starch absorption so as to achieve the effects of controlling postprandial blood sugar and controlling body weight, and is suitable for being used by general consumers. The composition specifically comprises the following raw materials in parts by weight: functional cereal powder, skimmed milk powder, inulin, resistant maltodextrin, walnut oligopeptide powder, pea oligopeptide powder, burdock powder, plant fat powder and a sweetening agent. By optimizing the method, modifying the starch, combining the starch with the beta-glucan, and combining with the spray drying technical means, the starch absorption is inhibited, the satiety is strong after eating, and meanwhile, the beta-glucan can accelerate the formation of mucus layers on small intestine walls and slow down the speed of conveying food to intestinal tracts, so that the digestion rate of the food is reduced, and the health of people is promoted. And finally, the postprandial blood sugar steady state can be maintained.

The application discloses a Bayer image denoising method, which comprises the following steps: obtaining a Bayer image and performing a demosaicing operation to obtain an RGB image; converting the format of the RGB image to obtain a YUV image; performing image denoising on the Y channel of the YUV image to obtain Ynr; performing downsampling on the Bayer image to obtain four chrominance subgraphs R1, Gr1, Gb1 and B1; performing downsampling on the Ynr image to obtain four luminance subgraphs YR1, YGr1, YGb1 and YB1; performing guided filtering on the chrominance subgraphs by using the luminance subgraphs to obtain four chrominance filtered images R1nr, Gr1nr, Gb1nr and B1nr; performing guided filtering on each level of the luminance subgraphs and the chrominance subgraphs by using a Gaussian pyramid; and performing upsampling and weighted fusion on each level of the four chrominance filtered images and restoring the images to the original positions to output a Bayer denoised image. The application takes into account low-frequency and high-frequency signals, improves the image signal-to-noise ratio, and retains more image details.

The invention relates to the technical field of fabric recovery, in particular to a production process of cyclically regenerated polylactic acid and low-temperature spandex fabric. The production process provided by the invention overcomes the problems of high stress of polylactic acid and low-temperature spandex fabrics at low temperature and poor dyeing and fixation effects. Four different types of fabrics including polylactic acid fibers, cotton fibers, polyester fibers and spandex are subjected to recovery treatment and physical treatment and then treated with an organic solvent and a phosphoric acid solution, and the raw materials are treated with a prepared PLA decomposition catalyst and a lactide ring-opening catalyst. Vitamin B12, a chain extender and isocyanate are supplemented in the technological process, and the dyeing process of low-temperature treatment is adopted, so that the stress of the regenerated polylactic acid and low-temperature spandex fabric is reduced, the dyeing and color fixing effects are good, and the closed loop of recycling and reusing of the polylactic acid and spandex waste fabric is realized.

The invention discloses a head and neck tumor image segmentation method based on combined analysis of CT and PET, and the method comprises the steps: obtaining a head and neck tumor data set, and carrying out the preprocessing of the head and neck tumor data set; constructing a parallel common attention mechanism module; an improved UNet3D network model is constructed; optimizing the improved UNet3D network model by adopting a combined loss function; and inputting the preprocessed head and neck tumor data set into the optimized and improved UNet3D network model, and outputting a segmentation task result of the bimodal image. According to the method, a parallel common attention mechanism and a multi-scale feature fusion technology are introduced, multi-modal image information is efficiently combined, features of the tumor under different scales are captured, accurate segmentation is carried out on the features, the precision and efficiency of head and neck tumor segmentation are remarkably improved, the limitation that a single-modal image is difficult to comprehensively reflect the tumor features is overcome, and the accuracy and efficiency of head and neck tumor segmentation are improved. And the problem of scale diversity of the tumor in the image is solved.

This invention discloses a virtual immunohistochemical staining method and system based on relation generative adversarial networks (GANs) and sequence slide data, comprising the following steps: Step S1, constructing a sequence slide dataset, wherein each H&E-stained slide is followed by three consecutive Ki67-stained slides; Step S2, constructing a relation generative adversarial network (GAN) model, the model comprising two ResNet generators and two relation discriminators, used to train and evaluate the credibility of the sequence relationship between image pairs; Step S3, performing two-stage training on the relation generative adversarial network model, namely, basic style pre-training and Z-axis alignment relation adaptation. This invention fundamentally solves the Z-axis alignment problem, resulting in higher image fidelity and pathological accuracy, significantly improving clinical application value, and providing new research resources and evaluation paradigms.

This invention discloses a specular removal method based on grouped enhanced convolutional attention feature fusion, belonging to the field of image processing technology. The method constructs a specular removal model based on a CycleGAN network, which consists of a generator, a discriminator, and a loss function optimization module. The generator includes detail enhancement depth downsampling, upsampling, and grouped enhanced convolutional attention fusion modules, effectively capturing image details, reducing checkerboard artifacts, and enhancing feature fusion and capture capabilities. The discriminator uses a PatchGAN structure to improve local detail capture capabilities. The optimized loss function integrates multiple losses to improve model performance. Experimental results show that compared with various traditional and deep learning methods, the method of this invention outperforms in PSNR and SSIM metrics, exhibits strong adaptability, and can provide high-quality image data for subsequent detection of elevator door components.

The invention discloses an image confrontation defense method and system based on halftone and a filter, and the method comprises the steps: carrying out the halftone processing of an original image I, obtaining a halftone image Ih, carrying out the filtering processing of the halftone image Ih through a filter, and obtaining a low-quality continuous tone image If; training an M-Net network by using the original image I and the low-quality continuous tone image If to obtain a high-quality image Ir; and performing halftone processing on the adversarial image A to obtain a halftone image Ah, performing filtering processing on the halftone image Ih by using a filter to obtain a low-quality continuous tone image Af, inputting a high-quality image Ir, and recovering the low-quality continuous tone image Af by using the trained M-Net network to obtain a high-quality image Ar without adversarial disturbance. The method can solve the problems that disturbance elimination of an existing confrontation and defense method is not thorough, and consequently the accuracy of a recognition model is reduced.

The invention relates to the technical field of energy knowledge maps, in particular to a comprehensive energy knowledge recommendation method and device, and the method comprises the steps: obtaining an energy information keyword in a question of a user, and obtaining a vector representation of the energy information keyword; forming the vectors of the energy information keywords into time sequence vectors according to a time sequence; according to the time sequence vector, predicting to obtain an interest keyword of the user; nodes similar to the interest keywords are determined in the comprehensive energy knowledge graph, and recommended content is determined. According to the recommendation method based on the comprehensive energy knowledge graph, personalized recommendation is rapidly and accurately performed on the user, the structure of the user and the comprehensive energy information can be reserved to the maximum extent, and then prediction and recommendation are accurately performed.

The application discloses a rumor detection framework based on adaptive data enhancement and adversarial training, and aims to overcome the problems that the prior art only considers structure and ignores features when facing data noise problems, and does not consider the influence of adversarial attacks on rumor detection problems. The steps of the method are as follows: 1. processing the data set; 2. using an adaptive data enhancement method to weaken the influence of noise; 3. using a graph convolutional neural network to generate a graph representation; 4. using an adversarial generation module to generate an adversarial representation to improve the robustness of the model; and 5. constructing a joint loss function to optimize the rumor detection task.

The present application relates to a kind of based on carbonized wood natural structure dopamine electrochemical sensor and its preparation method and application, the working electrode of the electrochemical sensor is modified with carbonized wood-ZnO nanowire-ZIF-8.The electrochemical sensor of the present application uses carbonized wood as substrate, after cutting carbonized wood whole piece, using conductive carbon paste is pasted on working electrode, while ensuring conductivity, also retain the three-dimensional structure of nanomaterial not be destroyed, guarantee the performance of sensor, with good application prospect.

The invention discloses an image migration method, an image migration device, image migration equipment, a storage medium and a computer program product, and relates to the technical field of computer vision. The invention provides a fine-grained image migration method based on a diffusion model. Semantic matching and appearance migration are carried out in a pixel space of a diffusion model, detail consistency between an original image and a target image is effectively kept, detail loss is avoided, and therefore image migration precision is improved; through a potential space deviation correction method based on feature offset processing, it is ensured that the structure and semantic features of the image are effectively reserved in the migration process, the high-fidelity image migration effect is achieved, and the structural integrity and the information fidelity are enhanced.

The present application relates to a kind of processing method for prescription data, electronic equipment and storage medium, it is related to data processing technical field, including the following steps, to electronic prescription data is carried out semantic entity extraction, obtain prescription sensitive data, and prescription sensitive data is carried out risk classification mark, obtain sensitive feature label;Prescription sensitive data is carried out random difference disturbance calculation based on sensitive feature label, obtain disturbance numerical sequence;Based on prescription sensitive data, disturbance numerical sequence is carried out feature cross and non-linear mapping, obtain confused prescription data;Confused prescription data is carried out node comparison based on pre-set drug knowledge graph, obtain graph matching node;Confused prescription data is carried out risk assessment and numerical update based on graph matching node, obtain target prescription data, it solves how to protect the technical problem of the premise of prescription data privacy, guarantee the availability of data to support risk assessment and knowledge graph association as main problem.

The invention relates to the field of data visualization, in particular to a scatter diagram visualization task-oriented data sampling quality evaluation method and system. The method comprises the following steps: giving a high-dimensional data set, and generating a high-dimensional data subset by using a data sampling method; obtaining a low-dimensional data subset according to the high-dimensional data subset; randomly extracting clusters of the high-dimensional data set to obtain high-dimensional space sub-clusters; obtaining a low-dimensional space sub-cluster according to the high-dimensional space sub-cluster; measuring the difference between the high-dimensional space sub-cluster and the low-dimensional space sub-cluster to obtain the difference between different dimensional space sub-cluster pairs; obtaining a global error according to the difference of the sub-cluster pairs in the different dimension spaces; and evaluating the sampling quality of the high-dimensional data through the global error. According to the method, by comparing the sub-clustering differences of different dimension spaces and measuring the retention conditions of different sampling methods after data projection, the authenticity of the high-dimensional data in the projection scatter diagram is effectively evaluated.

This invention relates to the field of polishing equipment technology, and discloses a multi-faceted polishing fixture for watch cases, including a processing box. A silicone mesh plate is located in the center of the inner side of the processing box, and a placement seat is provided on the silicone mesh plate. Multiple processing chambers for individually processing the watch case are equidistantly arranged on the placement seat. By adding and setting a front-end processing mechanism, which uses alkaline cleaning water immersion combined with ultrasonic vibration and a circulating filtration structure, various oil stains and impurities such as cutting fluid and release agents remaining on the surface and crevices of the watch case can be thoroughly removed, eliminating polishing slippage and surface dulling caused by oil residue. Secondly, relying on constant-temperature liquid immersion, the thin-walled watch case is heated evenly throughout, thoroughly and evenly releasing residual metal stress from previous processing, avoiding subsequent polishing deformation caused by uneven stress release. Simultaneously, it optimizes the polishing process, reduces grinding head vibration, and eliminates wavy lines and light / dark layer defects on the watch case surface.

The invention discloses a coronavirus appearance imitating hyperelastic memory alloy rod triggering type spherical robot. The robot is mainly composed of a rod end contact, a hyperelastic memory alloy rod, a tough wire, a rod sleeve, a spherical shell, a micro camera, a hexagonal connecting prism, a straight-through connector, a take-up and pay-off device, a dodecahedron inner core, a control panel and a nine-axis attitude sensor. The take-up and pay-off device drives a winding reel to rotate through a motor so as to take up the tough wire, an elastic rod on the outer surface of the spherical robot deforms, then the tough wire is rapidly released, the bent elastic rod rebounds to provide power for movement of the robot, meanwhile, the take-up and pay-off device and the elastic rod reset, and continuous movement of the robot is guaranteed. The bending degree of the elastic rod can be controlled by changing the position of the knot on the toughness line, and the ejection period is controlled by the control system. The robot is unique in structure, novel in principle, composite in function, high in man-machine interaction and sensing capacity and high in movement flexibility and stability, and has important theoretical significance and practical value for research of spherical robots.