96results about How to "Realize real-time detection" patented technology

The invention provides a construction method, a control method and a system of a flight control network, and the method comprises the steps: employing a depth deterministic strategy gradient algorithm, building an attitude control network of an unmanned aerial vehicle in a flight process based on the state information and a target flight value of the unmanned aerial vehicle in the flight process, employing a depth reinforcement learning algorithm, and carrying out the attitude control of the unmanned aerial vehicle in the flight process. Based on the state information and the information of the environment where the unmanned aerial vehicle is located, a path planning network and an intelligent avoidance network of the unmanned aerial vehicle in the flight process are built respectively; based on the attitude control network, the path planning network and the intelligent avoidance network, constructing a hierarchical deep reinforcement learning network of the unmanned aerial vehicle in the flight process; and training the hierarchical deep reinforcement learning network by adopting historical flight data of the unmanned aerial vehicle in different flight environments and a preset flight reward mechanism to obtain a flight control network of the unmanned aerial vehicle. According to the invention, the adaptability and flexibility of the unmanned aerial vehicle in the flight process can be improved.

The invention provides a rail transit foreign matter invasion risk identification and evaluation method and system, and belongs to the technical field of target identification of image processing. Image data of a running environment in front of a train are collected in real time; performing foreign matter detection on the acquired image data frame by frame through a foreign matter detection algorithm, and obtaining the category and position information of the detected foreign matter if the foreign matter is detected to exist in a certain frame of image; segmenting a track region in the frame of image through a track segmentation algorithm to obtain position information of the track region; based on the relative position relation between the foreign matter and the rail area, the distance between the foreign matter and the train and the category of the foreign matter, comprehensively evaluating and outputting the rail transit foreign matter invasion risk level. According to the method, the foreign matter invasion risk in the rail transit scene can be accurately identified in real time, meanwhile, a rail transit foreign matter invasion risk grade evaluation system is established, and efficient detection, identification and evaluation of the rail transit foreign matter invasion risk are achieved.

The invention relates to the technical field of quality management, in particular to a real-time quality management method and system for a PVC pipe, and the method can be applied to quality management of the PVC pipe, especially real-time defect detection of a high-speed production line. The method comprises the following steps: firstly, acquiring a multi-modal data set of the PVC pipe in real time and preprocessing the multi-modal data set to obtain preprocessed data; secondly, constructing a deep learning model, extracting multi-scale features of the preprocessed data by using depth separable convolution and asymmetric convolution, and reducing parameters of the deep learning model; thirdly, fusing the extracted multi-scale features by using multi-modal fusion, and carrying out feature dimension reduction in combination with global mean pooling; then, training a deep learning model and evaluating the detection performance thereof; and finally, detecting and alarming the surface defects of the PVC pipe in real time. According to the method, the surface defects of the PVC pipe are detected based on the deep learning model, and the detection accuracy is ensured while the real-time performance of detection is improved.

The invention discloses a fault classification method and device for arc current signals and a storage medium. The method comprises the following steps: collecting a current arc current signal, and preprocessing the current arc current signal to obtain a to-be-decomposed arc current signal; decomposing the arc current signal to be decomposed into a plurality of single fluctuation components based on a local fluctuation feature decomposition method, performing multi-dimensional feature data extraction on each single fluctuation component, and constructing a target feature vector corresponding to the single fluctuation component based on the extracted multi-dimensional features; and determining a fault classification result corresponding to the current arc current signal based on the target feature vectors corresponding to the multiple single fluctuation components and a pre-trained target signal classification model. Accurate fault identification of the arc current signal is realized, and the efficiency and accuracy of fault classification are improved.

The present invention belongs to the field of quantum navigation and positioning technology, and discloses an inter-satellite ranging system, method, device and data processing terminal. The inter-satellite ranging system includes: a photon transceiver unit, which is used to generate entangled photon pairs, transmit signal photons to the satellite to be measured and the internal optical path, and receive signal photons reflected by the satellite to be measured and the internal optical path; a coincidence counting unit, which is used to associate the received signal photons with the idle photons to obtain their second-order correlation curves; a distance measurement unit, which is used to perform DFT phase difference weighted processing on two second-order correlation curves generated by the optical path difference inside and outside the system to obtain the actual optical path difference from the satellite to be measured to the system. The present invention uses a photon matching method to filter out the cosmic space stray light entering the photon detector to a large extent, thereby improving the signal-to-noise ratio. The present invention adopts an improved coincidence counting algorithm, and the calculation rate is greatly improved compared with the traditional coincidence counting algorithm.

The invention belongs to the technical field of network security, and aims to provide an intelligent malicious code detection method and system based on a large language model. The method comprises the following steps: acquiring a pre-trained large language model, and taking the pre-trained large language model as an initial malicious code detection model; acquiring a malicious code detection sample task to perform parameter fine tuning on the initial malicious code detection model to obtain a fine-tuned malicious code detection model; acquiring a specified prompt template to finely adjust the fine-adjusted malicious code detection model to obtain a final malicious code detection model; and acquiring a to-be-detected malicious code and additional information thereof, and constructing according to the to-be-detected malicious code and the additional information thereof to obtain a malicious code detection task so as to obtain a malicious code detection result matched with the to-be-detected malicious code according to the final malicious code detection model. According to the method, the malicious code detection cost can be reduced, real-time detection of network attacks can be realized, and the detection rate is higher.

The invention discloses a motor shell surface defect detection method based on deep learning. The motor shell surface defect detection method comprises the following steps that S1, a data acquisition device acquires a defect image of a motor shell; s2, marking the defect image by a computer; s3, using a deep learning model to train data; and S4, outputting a detection result of the surface defects of the motor shell. The motor shell surface defect detection method based on deep learning disclosed by the invention has the beneficial effects that the deep learning model can process defect images at a high frame rate, real-time detection is realized, and the motor shell surface defect detection method is suitable for online detection application; through end-to-end model training, the deep learning method can automatically learn features from data, complex preprocessing and feature extraction steps are not needed, and the detection process is simplified.

The invention belongs to the technical field of tunnel detection, and particularly relates to a tunnel real-time detection method based on a mobile laser scanner, which comprises the following steps: step 1, laser scanner control and section data acquisition; 2, converting the coordinates of the scanning points; 3, real-time limit detection is carried out; and step 4, resolving structural deformation. According to the invention, tunnel geometric element measurement and limit detection are directly completed in the measurement process, real-time alarm is carried out in an area with problems, and workers can carry out rechecking and marking on site and directly carry out maintenance processing on the area with the problems; the tunnel three-dimensional point cloud, the tunnel laser intensity image, the tunnel detection report and other results can be generated in real time, the operation efficiency is greatly improved, and the workload of post-processing operation is reduced; and the tunnel intrusion condition can be marked in the image and the section point in real time, so that the field staff can position and process conveniently.

The invention relates to a carrier modulation method and an optical storage integrated system, and relates to the technical field of carrier modulation. A specific embodiment of the method comprises the following steps: respectively acquiring carrier signals of a plurality of DC / DC converters in the optical storage integrated system, taking one of the carrier signals as a reference carrier signal and the rest as carrier signals to be controlled, and calculating a carrier phase difference value between the carrier signals to be controlled and the reference carrier signal; and calculating an error value between the carrier phase difference value and a carrier phase preset difference value, and adjusting the carrier signal to be controlled according to the error value and an error value threshold value so as to realize carrier modulation on the DC / DC converter. According to the embodiment of the invention, automatic control of carrier wave interleaving parallel connection is realized, the situation of direct connection of a switching circuit in a carrier wave modulation process is avoided, the carrier wave modulation effect is improved, and the application scene of a carrier wave modulation scheme is expanded.

The embodiment of the invention relates to the field of batteries, and provides a method and equipment for automatically detecting and adjusting the weight loss ratio of a pole piece, and the method comprises the steps: providing drying equipment to dry the pole piece; providing a moisture content analyzer to measure the moisture content of the dried pole piece in real time; providing a programmable logic controller for obtaining the actual weight loss ratio of the dried pole piece based on the water content, setting a target weight loss ratio in the programmable logic controller, obtaining the actual error at the current moment based on the actual weight loss ratio and the target weight loss ratio, and if the actual error at the current moment meets a preset condition, sending the preset condition to the controller; the actual error serves as a current error, an expected error is obtained based on all the actual errors obtained within a previous preset time period including the current moment, an adjusting amount is obtained based on the current error and the expected error, and working parameters of the drying equipment are adjusted based on the adjusting amount so as to adjust the weight loss rate of the pole piece. The embodiment of the invention is at least beneficial to improving the qualified rate of the pole piece and reducing the preparation cost of the pole piece.

The utility model provides a gas detection device which mainly comprises a PWM (Pulse Width Modulation) speed regulation module, a vacuum self-priming pump, a gas chamber, a detection chamber comprising different gas sensors and signal conditioning circuits, a main control board, a stepping motor driving module, a stepping motor, an incremental encoder and the like which are fixed in a shell of the device, the parts can be physically and logically connected by adopting wires or communication protocols, the gas chamber can be respectively communicated with different detection chambers, gas to be detected can enter the different detection chambers through the gas chamber after the gas chamber is communicated, and the concentration of the gas is detected by utilizing different gas sensors and signal conditioning circuits; and a detection result is uploaded to the distributed control system, and alarm and real-time display are carried out through the sound-light alarm module and the display module. The real-time detection of different combustible and toxic gases is realized, the expansibility and universality of the gas detection device are greatly improved, the inventory occupation of power plant spare parts is reduced, and the economic benefit is improved.

The utility model discloses a bridge expansion joint displacement detection device and detection system, and the device comprises a detection rod and a protection sleeve with one end provided with an opening. One end of the detection rod is inserted into the protective sleeve in a sliding and sealing manner, a laser displacement sensor is fixedly mounted at the end, located in the protective sleeve, of the detection rod, and a reflection part right opposite to the laser displacement sensor is arranged in the protective sleeve; the end, away from the detection rod, of the protection sleeve is fixedly sleeved with a sleeve in a sealed mode, a first cavity is formed between the sleeve and the protection sleeve, a first through hole communicated with the first cavity is formed in the sleeve, and a second through hole communicated with the first cavity is formed in the protection sleeve. According to the utility model, the bridge expansion joint can be detected in real time, and the detection efficiency and precision are improved.

The invention relates to the technical field of sensor detection, and discloses a deep learning-based nitrogen-oxygen sensor aging test method, which comprises the following steps: extracting universal aging trend data from preprocessed historical large sample data; training the self-attention mechanism model by using the universal aging trend data to obtain a universal basic model; taking the trained general basic model as a transfer learning model, and performing model fine adjustment on the transfer learning model; adding a new full-connection layer, connecting the output of the fine-tuned transfer learning model to the new full-connection layer, activating the output of the new full-connection layer, and training the fine-tuned transfer learning model by using special aging trend data by taking minimization of a predefined loss function as a target to obtain a special detection model; performing an aging test on the to-be-detected nitrogen-oxygen sensor by using the special detection model; according to the invention, the accuracy and robustness of aging detection of the nitrogen-oxygen sensor can be improved.

The invention discloses a medicine dissolution rate testing device which comprises a base placed on the ground, a heating box is fixedly connected to the base, a plurality of evenly-distributed hole grooves are formed in the heating box, and the inner wall of each hole groove is movably connected with a dissolution cup for testing the medicine dissolution rate. The outer contour, close to the end part, of each dissolution cup is fixedly connected with a limiting ring for supporting and placing the dissolution cup on the inner wall of the hole groove, and one side of the base is fixedly connected with a rectangular shell through an external bracket; a stirring mechanism for stirring a solution in the dissolution cup and a liquid pumping mechanism for detecting the dissolution rate of a medicine in the dissolution cup are arranged on the rectangular shell, the device has the advantages that the medicine solution in the dissolution cup is quantitatively pumped and detected, and the dissolution rate testing precision is improved.

The invention discloses an intelligent monitoring system of electrical equipment, and relates to the technical field of electrical detection. According to the invention, through the fan blades and the conductive layer arranged in the air inlet pipe, real-time detection of the load current in the equipment can be realized by rotating the fan blades to cut a magnetic field. Meanwhile, the fan blades are matched with the first conducting layer and the second conducting layer, so that current can be effectively detected while air flow is maintained, and dual functions are achieved; a flow dividing channel and a separation chamber are arranged in the air inlet system, and through the exquisite design of an air suction nozzle and a separation pipe, air entering the equipment can be effectively filtered before entering the equipment. Dust and particulate matters in the air are separated and collected into the dust collecting groove under the action of centrifugal force, so that part of impurities entering the equipment are reduced, and the durability and the reliability of the equipment are further improved.

A weld defect detection method based on improved YOLOX belongs to the field of image recognition technology. This method adopts the currently widely used deep learning-based algorithm to realize the detection and identification of weld X-ray internal defects. The present invention is based on the YOLOX‑s algorithm, and adds an attention mechanism to its backbone feature extraction network Backbone, so that the model pays more attention to edge information. Secondly, some convolutions in the network are changed to deep separable convolutions to reduce the number of parameters. Finally, the cosine annealing method is used to adjust the learning rate in the model training process to achieve a better result. The method of the present invention can detect and identify internal defects of welds more quickly and accurately.

The present invention belongs to the technical field of optical detection in optical fiber transmission, and discloses an optical signal detection device applied to a bidirectional optical fiber transmission system, which solves the problem of dynamic optical power loss detection of optical signals in the bidirectional optical fiber transmission system composed of optical modules of the SFF‑8472 protocol standard during use. The device collects the real-time transmitted optical power and received optical power data of the optical modules in the bidirectional optical fiber transmission system, and calculates the dynamic change of the current optical power through the optical power threshold set by the host computer control computer; in addition, the device realizes optical power detection of up to 64 bidirectional optical fiber transmission systems. The present invention solves the problems of collection, calculation and result reporting of transmitted and received optical power in a bidirectional optical fiber transmission system, and has accurate optical power detection, which is conducive to locating optical path and circuit faults. It can also be used for dynamic real-time detection of optical power in a bidirectional optical fiber transmission system composed of optical modules of the SFF‑8472 protocol standard, and provides data support for it.

The invention provides a method for predicting coarse grain defects on the surface of a high-strength aluminum alloy extruded profile, and relates to the field of aluminum alloy image processing. The method for predicting the surface coarse grain defects of the high-strength aluminum alloy extruded profile comprises the following steps: carrying out a hot extrusion experiment and a hot compression experiment; representing an extruded profile and a thermal shrinkage pressure sample to obtain a recrystallization fraction and a recrystallization grain average size; simulating a profile extrusion process and obtaining basic physical quantities of the profile; determining a function relationship between the physical field quantity and the recrystallization fraction and between the physical field quantity and the recrystallization grain average size through data fitting and linear regression, and establishing a prediction mathematical model; establishing a recrystallization grain growth rate model; and realizing surface coarse grain defect prediction of the extruded profile through numerical simulation. According to the method, the distribution position of the coarse grain defects on the surface of the aluminum alloy profile can be accurately predicted, the recrystallization degree, the grain size and the recrystallization grain coarsening rate of each position of the profile are independently quantified, and the coarse grain defects on the surface of the high-strength aluminum alloy extruded profile are accurately evaluated.

The invention relates to a polarization-maintaining enhanced winding system and method for a hollow-core microstructure optical fiber ring. The polarization-maintaining enhanced winding system comprises a ring winding device and an optical fiber birefringence detection module. The winding equipment comprises two fiber releasing wheels, a winding framework and two groups of tension adjusting mechanisms; the optical fiber birefringence detection module comprises a wavelength tunable laser, a polarization state generator, a polarization state analyzer and an upper computer; the method comprises the following steps: starting an optical fiber birefringence detection module in the winding process of an optical fiber ring; the precise value of the optical fiber group delay inequality delta tau is obtained through an optical fiber birefringence detection module, and the birefringence value of the optical fiber is reversely deduced; the optical fiber birefringence is used as a feedback signal to control the tensile force applied to the hollow optical fiber by the tensile force adjusting mechanism, and the high-stability optical fiber birefringence is obtained. The polarization-maintaining enhancement effect can be achieved, and polarization-dependent noise in the interference type hollow-core microstructure fiber-optic gyroscope can be inhibited.

The utility model relates to an anti-collision detection structure and a washing machine. The anti-collision detection structure comprises a box body, an upper cover and a first motor, the upper cover is rotationally connected to the box body, and the first motor is arranged on the box body and can drive the upper cover to rotate; one of the box body and the upper cover is provided with a receiver, and the other one of the box body and the upper cover is provided with an emitter corresponding to the receiver; one of the receivers and the transmitter is further provided with a second motor, the second motor is electrically connected with the first motor, and the second motor can drive the receiver or the transmitter to rotate according to the rotation angle of the first motor, so that the receiver and the transmitter directly face each other all the time; according to the anti-collision detection structure, the emitter is driven by the second motor to rotate, so that the emitter always faces the receiver in the falling closing process of the upper cover, real-time detection in the falling closing process of the upper cover is achieved, the upper cover is prevented from hurting a user in the falling process, and the safety of the anti-collision detection structure is improved.

The invention discloses a fluorescent quantitative PCR (polymerase chain reaction) primer group for detecting paenibacillus mucilaginosus and application of the fluorescent quantitative PCR primer group. The sequence of the primer group is shown as SEQ ID NO.1-3, the reaction system comprises the primer group, a PCR buffer, Taq DNA polymerase, UDG enzyme, positive control and negative control, and the detection steps comprise: extracting nucleic acid of a sample to be detected, constructing the reaction system, carrying out a reaction according to a reaction program, and carrying out result interpretation. The method has the advantages that the detection method is efficient, simple and convenient, the specificity is high, the sensitivity can reach 10 < 2 > CFU / mL, and the detection result is accurate and reliable.

The utility model discloses a consumable detection device for a 3D printing extruder. The consumable detection device comprises a printing head shell, a wire extruding channel is vertically formed in the inner side of the printing head shell in a penetrating mode, and a detection channel, an adjusting groove and a square groove are formed in the inner side of the printing head shell. The inner wall of the detection channel is fixedly connected with a first fixing shaft and a clamping block. Through an innovative mechanical conduction structure, when the consumables are inserted from the top, the balls are pushed and eject the deflector rod, so that the distance between the magnet block on the deflector rod and the Hall sensor is changed, the entry state of the consumables is detected in real time, and meanwhile, the detection accuracy is improved. The Hall sensor can accurately sense the actual diameter of the consumables according to the difference of the shifting distance of the shifting rod caused by small change of the diameters of the consumables, real-time accurate control over the extrusion flow is achieved in combination with the driving motor and the transmission gear set, and the problem that in the prior art, the extrusion amount is unstable due to uneven diameters of the consumables is solved; and the consistency of printing quality is obviously improved.

The invention discloses industrial wastewater treatment equipment with a waste gas purification function, and relates to the technical field of wastewater treatment.The treatment equipment is used for treating impurities in wastewater and comprises a main body mechanism, an impurity separation mechanism, an impurity detection mechanism, a solution separation mechanism and a waste gas filtering mechanism, and the main body mechanism is connected with the impurity separation mechanism; the main body mechanism is connected with the impurity detection mechanism, the solution separation mechanism and the waste gas filtering mechanism, the impurity separation mechanism is connected with the solution separation mechanism, the main body mechanism comprises a mounting base and a reaction device, and the reaction device is arranged on the mounting base and comprises a reaction tank and a heating resistance wire. The reaction tank is fixedly connected with the mounting base, the heating resistance wire is arranged in the tank wall of the reaction tank and is connected with a power supply, and the mounting base is connected with the impurity separation mechanism.

The present invention discloses a method, system and medium for detecting corrosion defects of line hardware based on cloud-edge collaboration, and specifically relates to the field of defect detection technology. The method collects hardware corrosion defect data of transmission lines, and obtains defect samples and normal samples in the hardware corrosion defect data based on a detection and recognition model; the defect samples are input into an improved YOLOv8m model to train a new detection and recognition model; the detection and recognition model is updated with the new detection and recognition model to perform line hardware corrosion defect detection; in view of the problems that the corrosion shapes are changeable, the features are not obvious, and the key information is easily lost in the deep network, a method for detecting corrosion defects of transmission line hardware is proposed in combination with cloud-edge collaboration technology, which is used to supplement the corrosion features, reduce information loss, improve the detection accuracy of corrosion, realize data interaction between intelligent terminals and edge computing devices, and realize integrated operation and maintenance services from data collection on the transmission equipment side to the cloud platform.

The invention provides a screen projection picture anomaly detection method and device, equipment, a medium and a product. The method comprises the following steps: transmitting first current screen projection information to a screen projected node through at least one intermediate node; determining a first verification result corresponding to the first current frame data according to the position information; receiving a second verification result and a third verification result which are sent by the intermediate node and correspond to the first current screen projection information; and if it is determined based on the first verification result, the second verification result and the third verification result that data exception exists in the process of transmitting the first current frame data to the screen-projected node, and data exception also exists in the transmission process of frame data of a preset number of frames after the first current frame data, determining that the screen-projected picture is exceptional. According to the projection screen image abnormity detection method, real-time detection of projection screen image abnormity is realized.

The invention provides an ear return time delay detection method and device, electronic equipment and a storage medium, the method is used for detecting ear return time delay of an ear return system, the ear return system at least comprises an audio collector and an audio player, and first cache time delay is determined according to collection cache data volume in the audio collector; determining the time delay of the audio collector according to the first cache time delay and the hardware time delay of the audio collector; determining a second cache delay according to the playing cache data volume in the audio player; determining the time delay of the audio player according to the second cache time delay and the hardware time delay of the audio player; according to the time delay of the audio collector and the time delay of the audio player, determining the ear return time delay of the ear return system; by adopting the method provided by the invention, the cached data volume of the ear return system is utilized to carry out ear return delay detection, and external equipment does not need to be connected to carry out delay detection, so that the detection process of the ear return delay is simplified, and real-time calculation on the ear return delay of the ear return system is facilitated.

The present invention discloses a medium-low temperature geothermal resource exploration device, comprising a guide pulley arranged at a geothermal wellhead, a measuring rope wound on the guide pulley, and a winding device arranged beside the geothermal well for winding the measuring rope, one end of the measuring rope is connected to the winding device, and the other end extends into the geothermal well, and a measuring float is arranged at the end of the measuring rope located in the geothermal well; an adjustment component is arranged on the measuring rope, and the adjustment component comprises an outer cylinder, a piston rod, an end cover, a first clamping device and a second clamping device, the piston rod is in a T-shaped structure, the piston rod is slidably installed in the outer cylinder, the end cover is connected to the outer cylinder, the piston rod passes through the end cover and is hinged to the second clamping device, a weight sensor is arranged at the upper end of the end cover, a pressure plate is arranged in the outer cylinder, the pressure plate can slide on the piston rod, and the lower end of the pressure plate contacts the weight sensor. The device can realize the detection of water level in the geothermal well, and has the advantages of simple monitoring and not easy to damage.

The invention discloses an electric drive type comprehensive detection vehicle and a track inspection method, and the detection vehicle comprises a walking unit, and a control unit, a seat unit, a power supply unit, an overhead line system detection unit and a track inspection unit which can be quickly disassembled and assembled on the walking unit, the power supply unit is electrically connected with the walking unit, the control unit, the contact network detection unit and the track inspection unit and provides working power for all the units, the control unit is arranged at the front end of the walking unit and comprises a control board, an industrial personal computer and an electric control system, the industrial personal computer and the electric control system are integrated in the control board, and the industrial personal computer is electrically connected with the contact network detection unit and the track inspection unit. And the electric control system receives and processes the image data, is electrically connected with the walking unit, and controls the walking unit to work under the control of the control console. Modular design is adopted in the overall structure, assembly and disassembly of upper and lower tracks can be achieved in a short time, carrying is convenient, the applicable scene is expanded, rapid and real-time automatic non-contact detection is achieved, the overhauling efficiency is greatly improved, and safety is improved.

The utility model discloses a silicon controlled rectifier dimming control circuit, which relates to the technical field of illumination and comprises an AC (alternating current) input module, a power supply circuit module, a zero-cross detection module, a control signal input module, a digital signal processing module and a power switch filtering module. The input end of the AC input module is connected with an AC power supply, and the output end is connected with the power supply circuit module, the zero-cross detection module and the power switch filtering module. The power supply circuit module is used for supplying power to the digital signal processing module; the control signal input module is used for inputting a control signal into the digital signal processing module; the zero-crossing detection module is used for inputting a zero-crossing signal into the digital signal processing module; and the power switch filtering module is used for filtering the dimming signal output by the digital signal processing module. According to the utility model, double-control dimming can be realized.

The invention provides a plastic processing line transferring device and method.The line transferring device comprises a main conveying line mechanism, a first conveying line transferring assembly and a second conveying line transferring assembly, and the main conveying line mechanism comprises a front conveying line assembly and a rear conveying line assembly; the front conveying line assembly comprises a first conveying roller, a second conveying roller, a first conveying belt and a first installation support, the first conveying roller and the second conveying roller are installed on the first installation support, and the first conveying belt is installed outside the first conveying roller and the second conveying roller in an end-to-end mode. The rear conveying line assembly comprises a third conveying roller, a fourth conveying roller, a second conveying belt and a second mounting support, the third conveying roller and the fourth conveying roller are mounted on the second mounting support, and the second conveying belt is mounted outside the third conveying roller and the fourth conveying roller in an end-to-end connection mode. The appearance and the size of the plastic products can be detected in real time, line transferring and classified conveying can be automatically carried out, and the subsequent machining efficiency of the plastic products is improved.