100results about How to "Reduce blur" patented technology

The invention relates to the field of image enhancement, in particular to an image enhancement method and system of a vehicle-mounted image sensor. The method comprises the following steps: acquiring driving record images of a vehicle at multiple angles; dynamic scene self-adaptive exposure adjustment is carried out on the driving record images of the vehicle at the multiple angles, so that multiple self-adaptive exposure driving images are obtained; carrying out deep network noise identification on the plurality of adaptive exposure driving images, and carrying out dynamic filtering and noise reduction on sub-images one by one so as to generate a plurality of filtering and noise reduction optimized images; carrying out image texture analysis on the plurality of exposure sub-images, and carrying out texture distortion reconstruction to construct a plurality of texture reconstruction images; and road object visual identification is carried out on the plurality of texture reconstruction images, and adaptive contrast enhancement processing is carried out, so that a plurality of detail contrast enhancement images are constructed. According to the invention, a high-definition and stable panoramic vehicle-mounted image is realized.

The invention relates to the technical field of digital printing machines, in particular to an intelligent ink droplet control method and system of a digital printing machine and the digital printing machine. The method comprises the steps of real-time data acquisition, multi-dimensional parameter modeling, real-time ink droplet behavior prediction, self-adaptive ink jet parameter adjustment, closed-loop feedback control, medium characteristic self-adaptive matching, intelligent multi-region optimization and environmental factor compensation. Ink droplet behaviors and environmental data are collected in real time through a high-speed camera and various sensors, a nonlinear model is constructed by using a deep learning technology, and a dynamic parameter optimization curve is generated. According to the method, the printing quality and efficiency are remarkably improved, and the method is suitable for various media and complex scenes.

The invention discloses a virtual reality fusion method based on an unmanned aerial vehicle video, and relates to the technical field of virtual reality, and the method comprises the steps: predicting the current and next concern points of a user through eye movement tracking and Kalman filtering, preferentially distributing rendering resources to the areas, and employing a dynamic resource distribution mechanism to achieve the fusion of the current and next concern points of the user; according to the method, resources are allocated to a region about to be concerned by a user in advance, meanwhile, the processing priority of a secondary region is reduced, waste of computing resources is avoided, meanwhile, the moving track of a fixation point of the user is predicted through a Kalman filter, the next concern point of the user is rendered in advance, rendering of the concern region is completed when the visual angle is rapidly switched, and the user experience is improved. Compared with the prior art, the method has the advantages that image lagging and blurring felt by a user are reduced, rendering efficiency is greatly improved, user experience is smoother and more natural, a multi-level cache framework is adopted, the problem of loading delay in a traditional method is avoided, and the response speed of a system is effectively increased.

The invention discloses a remote sensing image semantic segmentation method, system and device and a medium, and the method comprises the following steps: obtaining remote sensing building data, and inputting the data into a pre-trained semantic segmentation network, so as to achieve the precise segmentation of a building in a remote sensing image. The semantic segmentation network includes an encoder module and a decoder module for adaptive context selection based on structure priori guidance. In the training stage, a special remote sensing building data set is utilized to train the network so as to optimize the segmentation performance of the network. According to the method, by introducing structure prior information and an adaptive context selection mechanism, the accuracy and efficiency of building segmentation in the remote sensing image are effectively improved, and the problem that remote sensing buildings are difficult to distinguish when pixel visual evidences are weak or the remote sensing buildings belong to small space structure objects is solved.

The invention relates to a physical and data driving combined fluid animation generation method and device, and the method comprises the steps: S1, building a transparent fluid motion field, applying consistency constraint to the optical flow of a bidirectional video sequence, carrying out the self-supervision training through the real fluid video data in an original video data set S, and obtaining a transparent fluid motion field; generating average optical flow data F and constructing training data; s2, an animation generation model training stage: inputting the training data into a multi-scale double-flow texture feature learning network based on self-attention driving, and performing training in combination with an original video data set S and average optical flow data F; and S3, a fluid animation generation stage: simulating an input image by using a physical model to obtain a motion field M, and guiding actual natural fluid animation generation. According to the method provided by the invention, the physical authenticity of fluid animation generation is improved, and a new solution is provided for simulation and visualization of a complex fluid phenomenon.

The invention proposes a small sample image generation method based on semantic mask reconstruction and category semantic learning, and relates to the technical field of image processing, and the method comprises the steps: S1, constructing a network model comprising a generator and a discriminator based on a LoFGAN network; s2, extracting depth features of the input image by using a generator, and fusing the depth features of the input image into depth semantic features through local feature fusion; s3, semantic mask reconstruction is carried out on the depth semantic features, and a corresponding image is generated through a decoder; s4, sending a real image and a corresponding image generated by the decoder into a discriminator to train the network model; s5, inputting the small sample image into the trained network model to obtain a generated image similar to the input image; according to the method, the extraction precision of the semantic information can be effectively enhanced, the understanding ability of the generation model for a complex semantic scene is improved, and the fidelity and definition of the image are improved.

The invention relates to the technical field of machine vision, in particular to an image segmentation method and system for liquid crystal screen display. The method comprises the following steps: acquiring basic data of the liquid crystal screen; performing screen response characteristic analysis according to the basic data of the liquid crystal screen to obtain screen response characteristic data; performing multi-scale image acquisition according to the basic data of the liquid crystal screen to generate multi-scale image data; performing characteristic classification on the liquid crystal display according to the screen response characteristic data, and performing edge compensation analysis on the multi-scale image data to obtain edge characteristic compensation data; acquiring real-time data of the touch screen; and performing touch interference identification analysis on the real-time data of the touch screen according to the edge feature compensation data to generate touch interference mask data. By analyzing the screen response characteristics and performing edge compensation analysis, the phenomenon of edge blurring caused by liquid crystal response time limitation is reduced, and the accuracy of edge detection is improved.

The invention relates to the field of automatic driving scene reconstruction, discloses an automatic driving boundary-free scene implicit reconstruction method and system considering geometric information enhancement, and adopts an automatic driving geometric perception grid-based neural rendering system. The system reconstructs and renders an accurate driving environment from sparse sensor data using a perspective deformed hash grid and a signed distance function (SDF). The system comprises a Hash pool, 16 levels are included, 219 two-dimensional feature vectors are held in each level, the feature vectors are input into an MLP network through spatial interpolation, scene features and SDF are extracted, finally, high-quality scene reconstruction is achieved through optimization of RGB images, depth and normal, and RGB input and monocular depth and normal observation are used as supervision during joint optimization. According to the system, the scene reconstruction quality is improved, a boundless scene can be effectively processed, the performance is improved in a viewpoint sparse environment, and higher robustness and accuracy are shown when a complex scene in an actual driving environment is processed.

The invention belongs to the technical field of photoinitiators, and particularly relates to a ketoxime ester photoinitiator as well as application and a photosensitive combination thereof. Compared with the prior art, the novel photoinitiator has the advantages that by introducing a sulfone phenyl specific structure to a ketoxime ester group and the like, excellent photosensitivity can be shown in the optical wavelength range of 280-400 nanometers, and compared with OXE-01 and OXE-02 commonly used in the market, the novel photoinitiator can show better photoinitiation efficiency under an ultraviolet visible (UV-Vi s) LED light source.

The utility model relates to the technical field of electromechanical engineering construction, and discloses an on-site monitoring device for electromechanical engineering construction, which comprises a camera stabilizing platform and a movable base, the camera stabilizing platform is positioned above the movable base, and the bottom of the camera stabilizing platform is fixedly connected with three upper hinge seats. And the three upper hinge seats are circumferentially arranged relative to the bottom of the camera stabilizing platform. According to the on-site monitoring device for electromechanical engineering construction, through synchronous expansion or contraction of a sliding roller, a sliding block and a lower hinge seat and rotation of a connecting rod and a linkage rod, accurate control over the height of a camera stabilizing platform is achieved, the height of a camera can be changed, and the situation that a shot picture is blurred due to the fact that supporting legs are adjusted one by one is avoided; when a camera or equipment is fixed on the camera mounting platform, the camera or equipment can be kept stable in the horizontal direction, vibration and shaking caused by inclination or instability due to movement can be reduced, and clearer and more stable shot images can be obtained.

The invention discloses an image processing method and device, an image encoding and decoding system, equipment and a storage medium. The image processing method comprises the following steps: decoding a text code stream from a coding end through a first decoder to obtain semantic information of an original image; decoding a feature code stream from a coding end through a second decoder to obtain feature decoding information of the original image; processing the feature decoding information through a texture information acquisition module to obtain texture information of the original image; processing the semantic information and the texture information based on a hidden layer representation diffusion model to obtain fusion information; and processing the fusion information through an image generation module to obtain a decoded image. According to the image processing method provided by the embodiment of the invention, the semantic information and the texture information are processed based on the hidden layer representation diffusion model to obtain the fusion information, and the fusion information is processed through the image generation module to obtain the decoded image, so that a relatively clear image can be obtained, the condition of image blurring is greatly reduced, and the processing effect is relatively good.

The invention relates to a lightweight semantic segmentation method for strawberry flower and fruit recognition in a complex scene, and the method comprises the following steps: obtaining a data set of strawberry images, and carrying out the image collection of strawberry flowers and fruits in a scene covering different illumination, time and plant states; carrying out data processing, and carrying out classification and mask labeling on the screened strawberry images; a deep and narrow backbone network is designed, and data features of rich and complex strawberry images are obtained; introducing more attention to the boundary of the strawberry image through an edge attention mechanism guiding model to obtain a clear edge; a pyramid pooling module is introduced at a de-wharf, and global information is perceived by aggregating strawberry image features of different scales; the strawberry image is segmented through the convolutional neural network, so that strawberry image segmentation in a complex environment is dealt with, and the lightweight and accuracy of the model are improved.

The embodiment of the invention provides an optical module and optical display equipment. Wherein the optical module comprises a lens group, a light splitting element, a phase retarder and a polarization reflection element which are arranged along the same optical axis, and the phase retarder is located between the light splitting element and the polarization reflection element; the lens group at least comprises two lenses; the lens group is provided with at least one lens with negative focal power, and the lens with the negative focal power meets the condition that (T1 / T2) * (v-50) / n is more than or equal to 3 and less than or equal to 17; wherein T1 is the thinnest thickness of the lens with the negative focal power in the optical effective diameter, T2 is the maximum thickness of the lens with the negative focal power in the optical effective diameter, and n and v are the refractive index and the Abbe number of the lens with the negative focal power respectively.

The invention discloses early warning equipment for water conservancy flood prevention and an early warning method thereof, and relates to the technical field of alarm devices. Converting the water gauge image into an HSV color space to obtain an HSV image; carrying out hue, saturation and brightness separation on the HSV image, and respectively carrying out graying processing on separated components; whether the water gauge image belongs to a low-light image or a normal-light image is judged through the brightness component; if the image is a low-light image, brightness enhancement processing is carried out; carrying out image binarization processing and filtering processing; calculating a minimum surrounding matrix of a water gauge for the filtered image, and extracting a water gauge area; counting the number of scale marks in the water gauge area, and calculating to obtain a water level value according to the number; if the water level value is larger than the warning value, an alarm instruction is sent. A water level value is calculated by processing a low-illumination water level gauge image and accurately extracting a water level gauge, and early warning information is sent according to the water level value.

The invention relates to the technical field of unmanned aerial vehicles, and discloses a building deformation detection device based on unmanned aerial vehicle vision, which comprises a surveying and mapping unmanned aerial vehicle, a surveying and mapping camera is mounted at the belly of the surveying and mapping unmanned aerial vehicle, and the surveying and mapping surface of the surveying and mapping camera is divided into a radar surveying and mapping surface and an image photographing surface and is provided with a stage surface cleaning mechanism. The surveying and mapping surface is used for dynamically managing the surveying and mapping camera; the first protective film is timely wound and replaced through the winding piece, it can be ensured that the radar surveying and mapping part is always in the optimal working state, the first protective film can effectively prevent the radar surveying and mapping part from being polluted and damaged by external environments such as moisture, dust and dirt, and therefore it is ensured that the radar can continuously and stably work; and the data accuracy and reliability are improved.

The invention relates to a self-adaptive vibration reduction support for a vision measurement system. The vibration reduction support comprises a lower base and an upper supporting plate and is used for placing vision measurement equipment and a filtering device. A loading shock isolation assembly is arranged in the middle of the vibration reduction support, and the three-direction vibration isolation design is achieved through the synergistic effect of the evenly-distributed vertical main spiral springs and the quasi-zero stiffness shock isolation device. According to the shock isolation device, a horizontal rigid supporting base forms transverse constraint on a bistable beam, fastening connection is achieved through screw rod design, and the quasi-zero stiffness characteristic is jointly formed through the horizontal rigid supporting base and a lower auxiliary spiral spring. A special bistable beam elastic structure is designed, so that a bearing effect is provided for upper equipment, and a constant three-way vibration isolation effect can be kept under different frequencies. And the upper digital filtering device is used for further processing high-frequency noise and ensuring the measurement precision and stability of the vision measurement system. According to the self-adaptive vibration reduction support, digital filtering and physical vibration isolation are fused, and a complete set of method for improving the vision measurement precision is achieved.

The invention discloses a brain connection event causal relationship extraction method based on weighted joint constraint learning. The method aims at a text mining task-brain connection event causal relationship extraction in an event-level neural image. According to the method, a novel model named as a brain connection event relation miner is developed by utilizing weighted joint constraint learning. A lightweight model is developed by integrating a BiLSTM (Bidirectional Long Short Term Memory) network and a multilayer perceptron (MLP), and is used for jointly extracting a plurality of event causal relationships from brain connection literatures. In view of scarcity of an annotation brain connection corpus, a weighted joint constraint learning framework is provided. The framework integrates dual consistency constraints, including common sense constraints and domain constraints, and combines the dual consistency constraints with adaptive weight learning to enhance the small sample learning ability of the model.

The invention discloses a high-resolution remote sensing image compression method and system based on block modulation imaging, a binary mask matrix is generated, an acquisition end configures an optical mask device according to the binary mask matrix, and when a high-resolution remote sensing image is acquired, an optical lens is adopted to acquire an optical signal matrix; after optical mask processing is carried out through the optical mask device, modulation is carried out through the relay lens, then the optical mask device is converted into an electrical signal matrix through the photoelectric sensor, then the electrical signal matrix is partitioned and superposed so as to complete compression, and an obtained measurement value matrix is sent to a receiving end; the receiving end extracts a block mask matrix corresponding to each block from the binary mask matrix according to the block mode of the acquisition end, then extracts each block signal matrix from the measured value matrix to form a three-dimensional signal matrix, and inputs the three-dimensional signal matrix into a pre-constructed and trained decoding network; and decoding to obtain an optical signal matrix of the high-resolution remote sensing image. According to the invention, block modulation imaging is provided, and the decoding restoration quality is improved while the compression speed is improved.

The invention discloses liquid crystal display screen high-precision AOI online detection integral equipment, and relates to the technical field of liquid crystal display screen online detection, the liquid crystal display screen high-precision AOI online detection integral equipment comprises a machine body, supporting legs are fixedly mounted at the bottom of the machine body, detection equipment is fixedly mounted at the top of the inner wall of the machine body, and conveying wheels are rotatably mounted at the bottom of the inner wall of the machine body; a motor is fixedly installed at the bottom of the inner wall of the machine body, conveying wheels are fixedly installed at the output end of the motor and are in transmission connection through a transmission belt, a bulldozing device is arranged at the bottom of the inner wall of the machine body, a cleaning device is arranged at the bottom of the inner wall of the machine body, and a buffering device is arranged at the bottom of the inner wall of the machine body. The bulldozing device comprises a first gear, a rotating column, a second gear, a cylinder, a lantern ring and a stop lever, a sliding rod slides to push a push plate to start to move, a display screen is located in the middle of a conveying wheel, it is ensured that the measuring instrument is within the optimal measuring range, and the overall efficiency of detection is improved.

The utility model discloses an artificial intelligence video acquisition device, comprising a video acquisition processor, the front side of the video acquisition processor is provided with a lifting adjusting mechanism, through the arranged lifting adjusting mechanism, in the process that a video camera shoots and acquires a video, a rotating lead screw is controlled to rotate through a motor, and when the rotating lead screw rotates, the rotating lead screw is driven to rotate; the lifting sliding block in threaded connection with the outer side of the rotating lead screw can move in the vertical direction together with the video camera, so that the video camera can be adjusted to different heights according to shooting requirements, and richer visual angles and a larger shooting range are obtained; the camera is kept stable in the moving process, shaking and blurring caused by movement are reduced, the camera is particularly suitable for scenes needing long-time stable acquisition and shooting, and in addition, the video camera is covered and shielded through the protective cover, light can be effectively shielded, and the light is prevented from influencing the image acquisition of the video camera.

The invention provides a method for automatically focusing image definition by a three-dimensional endoscope camera, which comprises the following steps of: S1, determining initial positions of a left lens and a right lens of the three-dimensional endoscope camera, and setting an effective process range of a focusing motor; s2, within the effective process range of the focusing motor, taking at least one path of the three-dimensional endoscope camera as a criterion, and using a square gradient as effective measurement to quantify the definition of the current position of the image; s3, determining a peak position with the highest image definition by adopting a two-step search strategy; and S4, repeating the steps S2 to S3 until the left lens and the right lens of the three-dimensional endoscope camera reach a focusing success state. The invention can effectively solve the problems that the lens moves and the image cannot be clearly imaged in the later stage of the equipment, thereby further improving the utilization rate of the equipment and reducing the medical cost.

The invention discloses a residual phase error correction coefficient estimation method based on a sub-aperture cancellation joint quadratic fitting method, and the method comprises the following steps: (1), preliminarily estimating a radar speed approximate value through employing a satellite orbit speed and a wave beam moving speed, and carrying out the rough imaging of SAR echo data through employing a wKA algorithm, and generating a complex image; (2) windowing interception is carried out on the complex image, an initial correction parameter is set, sub-aperture segmentation and preprocessing are carried out, an objective function E1 (X) is constructed, and an image difference value after sub-aperture imaging is calculated; (3) gradually approaching the optimal solution of the phase error through a reflection, expansion and contraction optimization strategy; (4) after a termination condition is reached, constructing a target function # imgabs0 #, and performing fine optimization based on quadratic fitting to obtain an optimal phase correction value x; (5) carrying out azimuth compression on the obtained optimal phase correction value x by using a wKA algorithm to realize azimuth imaging; the method can stably work in different scenes and radar configurations.

The present invention relates to the technical field of electronic non-stop toll collection, and in particular to an electronic non-stop toll collection system based on high-precision identification, including a data acquisition module, an information processing module, and an identification and deduction module. The present invention greatly improves the accuracy of license plate recognition and the reliability of the system through advanced data acquisition and processing technology. The system collects license plate images through image sensors in the environmental monitoring area, and uses character recognition and judgment units for high-precision recognition to ensure that a high recognition rate can be maintained even under adverse weather conditions. By real-time monitoring of vehicle speed and image clarity, the system can dynamically adjust the heating temperature to prevent image blur and equipment failure. The system automatically adjusts system parameters for interfering weather such as rain and snow, thereby improving the stability of license plate recognition under various climatic conditions. It effectively solves the problem of low accuracy of license plate information recognition under extreme weather conditions due to reliance on light sources and light sensors.

The utility model discloses a high composite borosilicate glass window, which relates to the technical field of glass windows and comprises an explosion-proof membrane, a glass layer is fixedly connected to the inner surface of one side of the explosion-proof membrane, and a vacuum layer is arranged inside one side of the glass layer. The rainwater can be quickly guided by the water guide pieces, flows in the inclined direction of the water guide pieces and enters the drainage groove through the communicating groove along with flowing of the rainwater, the drainage groove is formed in the inner wall of the sealing ring so that the rainwater flowing down from the water guide pieces can be collected, and then the rainwater in the drainage groove can enter the drainage groove through the leakage holes. The conical design of the leakage holes can further ensure that rainwater flows out smoothly and cannot be accumulated in the drainage groove, the drainage groove inclines inwards by 15 degrees, the rainwater can quickly flow to the water collecting pipe at the bottom, the rainwater is prevented from being accumulated on the surface of the glass window through the structure, and the water stain and blurring phenomena are reduced.

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 present invention discloses a tunnel lining inspection climbing robot, which relates to the technical field of robot tunnel inspection, and comprises: a robot and wheels rotatably connected to the bottom surface of the robot, an adsorption cavity opening is arranged at the center of the bottom surface of the robot, a manipulator is installed on the upper surface of the robot, a camera is fixedly connected to the execution end of the manipulator, a wall cleaning mechanism is arranged on the side of the robot away from the manipulator, impurities such as moss and dust on the lining wall are removed in advance by a bucket, and these impurities are avoided from being sucked into the adsorption cavity opening, thereby reducing the possibility of blockage, which not only ensures the negative pressure adsorption effect of the adsorption cavity opening, so that the robot can be adsorbed on the wall more stably, but also avoids the situation that the robot slides down and falls and is damaged due to insufficient adsorption force, thereby saving time and energy, thereby improving the overall efficiency of the inspection work.

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 an intelligent traffic light identification and real-time prompting system for color blindness drivers. The system comprises an environment sensing module; the acquisition module is responsible for acquiring surrounding environment information; the parameter self-adaptive module is responsible for dynamically adjusting identification parameters of the system; the dynamic image processing module is responsible for processing and analyzing images acquired in real time; the signal lamp identification module is responsible for identifying colors and states of traffic lights; the decision and prompt module is responsible for making a corresponding driving decision and prompting a driver; according to the system, through the environment sensing module, the system can obtain and process the information of the surrounding environment in real time, the information comprises illumination conditions, weather conditions and the like, and basic data is provided for subsequent parameter adjustment and image processing. And the parameter adaptive module dynamically adjusts image processing parameters and signal lamp identification parameters according to the environment information, so that the adaptability of the system to different environments is improved.

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.