77results about How to "Avoid missed detection" patented technology

The invention, which relates to the intelligent traffic industry, discloses a high efficient and real-time analytical method of vehicle fake license suspicion and an apparatus thereof. According to the invention, all data for fake license analyses are stored in a memory; it is maintained that each vehicle is corresponded to only one record during the processing process and the total amount of the data is controlled; and it is determined whether a vehicle a fake licensed one by calculating whether a time difference of the same vehicle between two checkposts is less than a time threshold between the two checkposts, so that it can be rapidly and accurately determined whether there is fake license suspicion. Because distance between checkposts are various and the speed limits are different, time for driving a distance between each two checkposts is different from others; therefore, the shortest time for passing through the two checkposts can be determined by calculating a time threshold between the two checkposts; and the determination result is more reasonable than an arranged unified global time threshold. Moreover, a memory analytical mode is used to replace a traditional Oracle database query analytical mode; therefore, efficiency is high and operability is strong; and an operation burden of the Oracle database is reduced.

The invention relates to a sheet workpiece phased array ultrasonic guided wave imaging detection system and a detection method of the system. The system comprises a phased array ultrasonic detector, a phased array ultrasonic guided wave probe and a position encoder. The phased array ultrasonic guided wave probe comprises a phased array ultrasonic probe body and an ultrasonic guided wave wedge, an inclined rectangular groove is formed in the vertical side face of the ultrasonic guided wave wedge, the phased array ultrasonic probe body further comprises a probe main body and array wafers, the phased array ultrasonic probe body is installed on the inclined face of the ultrasonic guided wave wedge, the phased array ultrasonic probe body is tightly combined with the ultrasonic guided wave wedge through a coupling agent, and the array wafers are distributed on the inclined face in the depth direction of the inclined rectangular groove. The ultrasonic guided wave wedge and the array wafers are adopted for large-area detection, so that detection efficiency is improved, edges of detection areas are covered with each other in an overlapped mode, and detection results are more accurate; the inclined rectangular groove is formed to generate a coupling monitoring signal and a detection signal which are matched with each other so that defects of sheet workpieces can be judged, and accuracy of the detection results is further improved.

The invention relates to a road traffic sign automatic detection and identification method. The method is realized based on an automatic detection and identification system. The automatic detection and identification system comprises a traffic sign acquisition module, a traffic sign detection and processing module and a traffic sign identification module. The traffic sign acquisition module is used for collecting driving videos; the traffic sign detection and processing module is used for carrying out processing on the collected videos, detecting image frames containing traffic signs and carrying out extraction on regions of interest; and the traffic sign identification module is used for carrying out classification and recognition on images of the extracted regions of interest. The road traffic sign automatic detection and identification method can identify various types of traffic signs, is high in precision, high in identification accuracy and good in robustness, reduces influence of illumination, geometric deformation and rotation and the like, and has better anti-interference capability.

The invention relates to an on-vehicle device and a recognition support system, capable of achieving cost reduction of the on-vehicle device and allowing a driver to reliably recognize the presence of a detected moving object while preventing detection omission of a moving object approaching an own vehicle from a blind corner for the driver. The on-vehicle device includes an imaging unit that captures an image of a wide angle range through a curved light receiver, a moving-object detector that detects a moving object approaching an own vehicle from a mounting direction of the imaging unit and from any direction other than the mounting direction, based on the image captured by the imaging unit, a collision-risk determination unit that determines a degree of collision risk indicating a possibility of collision between the moving object detected by the moving-object detector and the own vehicle, and a moving-object warning unit that warns presence of the moving object according to the degree of collision risk .

The present invention provides a target recognition method, system and device in a commodity image, and a storage medium. The method comprises: S1: acquiring a plurality of images, inputting the images into a pre-trained scenario classification model, and determining an application scenario corresponding to an image; S2: selecting a pre-trained ROI area detection model according to the applicationscenario corresponding to the image, and detecting ROI areas on the image; S3: correcting the image with the detected ROI areas to generate a corrected image; and S4: inputting the corrected image into a pre-trained target recognition model to perform recognition of each commodity area on the image. According to the technical scheme of the present invention, before the recognition of the commodity area, scenario classification, quality adjustment, ROI area recognition, and image correction are sequentially performed in an innovative manner, and finally recognition of the commodity area is performed, so that the recognition accuracy of the commodity area is remarkably improved, the recognition accuracy of the commodity is increased from 60% to 95%, and the application of the commodity recognition technology in new retail areas is promoted.

The invention discloses automatic detecting and adding methods and a system for a base station, a terminal and an adjacent region relation. The methods comprise the following steps of: sending a detection instruction containing preset detection frequency points to the terminal through the base station, then, detecting cells on the preset detection frequency points according to the detection instruction by the terminal, feeding back cell information of the cells to the base station, wherein detected signal intensity of the cell information can meet a preset condition, and the base station can judge whether a new adjacent region exists or not according to the cell information. Namely, when the invention detects an adjacent region, the cell information of all cells which are detected on the detection frequency points are really not sent to the base station, but the cell information of the cells is sent to the base station, wherein the signal intensity of the cell information can meet the preset condition. Therefore, the detected cells can be effectively screened, redundant information in an adjacent region relation allocation process can be greatly reduced, an adjacent region relation maintenance process can be simplified, adjacent region relation maintenance efficiency can be improved, moreover, when the terminal is switched, the adjacent region to be switched can be directly found as soon as possible, and efficiency and a success ratio for terminal switching can be improved.

The present application relates to a black screen detection method, a device, a computer device and a storage medium. The method comprises the following steps: when a video player is in a playing state, collecting picture frames being played in a playing interface of the video player according to a specified sampling frequency; acquiring similarity indication information between the at least two collected picture frames; When a degree of similarity between at least two picture frames indicated by the similarity indication information satisfies a preset condition, determining that a playing interface of the video player generates a black screen. In this process, since at least two picture frames compared are picture frames collected from the same playing interface, when the two picture frames are both picture frames of a black screen picture, sufficient consistency can be ensured between the two picture frames, leakage detection of the black screen can be effectively avoided, and the effect of improving the accuracy of the black screen detection can be achieved.

The embodiment of the present application discloses a method, a device, a storage medium and a terminal for detecting abnormal boot-up of a terminal. The method includes detecting an init process startup; obtaining the state information of the boot animation process through the preset detection service; when the startup failure is determined according to the state information, saving the target information causing the startup failure fault in the startup process. Through the technical proposal, the preset detection service can be started after the boot process enters the Android layer to detect the boot state, and the log causing the non-boot problem in the whole process can be obtained and stored when the boot fails; It can cover the entire phase of the boot process to avoid leakage detection problems. In addition, it can save the target information that causes the boot failure, and also facilitate the terminal developers to analyze the causes of non-boot problems.

The invention provides a human body sensing device and image forming apparatus having the same, being able to highly accurately determine the position and moving state of a human body. The human body sensing device comprises: an infrared sensor which detects a human body; a rotatable group of lenses having a plurality of lenses arranged around the sensor, the lenses being configured to: generate unit sensing zones defined by a sensing distance extending radially outwardly from the group and a small width extending in a rotation direction of the group; and form a plurality of alternating layers of a regional sensing zone including one or more than one of the unit sensing zones and a non-sensing zone not including any of the unit sensing zones, in a rotation direction of the group; a rotary drive which rotates the group; a rotation position sensor which detects a rotation position of the group; and a recognizer which recognizes the location and / or movement of a human body based on the output level indicated by the infrared sensor and the rotation position.

The embodiments of the invention disclose a chip and a detection method thereof. The chip includes a substrate, at least one laser detection circuit, and multiple key circuit modules. The laser detection circuit includes laser detection units, a NAND logic unit, and an AND logic unit. Each laser detection unit includes a monitoring diode and a first charging transistor; the anode of the monitoring diode is grounded; the input end of the first charging transistor is electrically connected with a power supply; and the cathode of the monitoring diode and the output end of the first charging transistor are electrically connected and serve as the output end of the laser detection unit. Each input end of the NAND logic unit is electrically connected with the output end of one laser detection unit, a first input end of the AND logic unit is electrically connected with the output end of the NAND logic unit, and a second input end of the AND logic unit receives an enable signal. Each key circuit module is set adjacent to at least one laser detection unit. Through the implementation of the embodiments of the invention, the accuracy of laser detection is improved, and missing detection is avoided.

The invention discloses an abnormality detection method and a device for a beam position monitor based on an auto-encoder. The method comprises: obtaining turn-by-turn beam position data of the beam position monitor; inputting the turn-by-turn beam position data as raw input data into a pre-trained auto-encoder model after preprocessing, processing and outputting reconstructed input data by the auto-encoder model; calculating a reconstruction error between the raw input data and the reconstructed input data; and determining whether the corresponding beam position monitor is abnormal by comparing the reconstruction error with an abnormality threshold, wherein the beam position monitor whose reconstruction error exceeds the threshold is determined to be in an abnormal or fault state. The abnormality detecting method and the device for the beam position monitor based on the auto-encoder solves the problems of abnormality of beam position monitors of storage rings and automatic identification of faults, improves the efficiency and reduces the research and development cost, and the method has high accuracy, high reliability and high practicability.

The invention relates to a detection device, and particularly relates to a lithium battery voltage detection device. The invention aims to provide a lithium battery voltage detection device. In orderto realize the purpose, the invention provides the lithium battery voltage detection device. The lithium battery voltage detection device comprises a fixed frame, a first wire, a second wire, an adjusting device, a voltmeter, a first contact, a second contact, a sliding rod, a sliding sleeve, a placement frame, a second sliding block and the like. The voltmeter is installed on the left side of thefixed frame. A second sliding rail is installed on the right side of the upper portion of the fixed frame. The second sliding block and a third sliding block are arranged on the second sliding rail in a sliding mode. The right part of the second sliding block is provided with the first contact. The right part of the third sliding block is provided with the second contact. The first contact is connected with the voltmeter through the first wire. The device is fast in detection speed, labor-saving, and high in detection efficiency. By means of the device, the voltage detection work of a large amount of lithium batteries can be completed while only a small amount of manual labor is required.

A weft knitting machine can detect abnormal vibrations resulting from an abnormal contact between a carriage and knitting needles, in a high sensitivity. A method can detect those abnormal vibrations. The weft knitting machine comprises a needle bed (1) having a plurality of knitting needles arrayed in parallel, and a carriage (2) running reciprocally over the needle bed (1) for moving the knitting needles back and forth. The carriage (2) includes vibration detecting means (or a vibration sensor (4)), which detects the vibrations to occur in a direction to intersect the running direction of the carriage (2). On the basis of the detection result, abnormal deciding means decides whether or not the abnormal vibrations are caused by the abnormal contact between the carriage (2) and the knitting needles. Even in case the running speed of the carriage (2) is high, the abnormal vibrations can be so detected as are discriminated from the vibrations accompanying the normal knitting operation.

The invention discloses a hardware authorization method, which comprises a production test step and a hardware use step; the production test step comprises the following steps: S11: collecting hardware attribute information to generate a machine code, the hardware attribute information comprising a hardware production date, a hardware serial number and hardware model information; s12, sending the machine code to a server for verification, specifically, enabling the server to compare the machine code with a to-be-authorized machine code list pre-stored on the machine code, and verification is passed if the machine code is in the to-be-authorized list; s13, after the verification of the machine code is passed, enabling the server to encrypt the machine code to form an encrypted character string, and sending the encrypted character string to hardware; s14, enabling the hardware to store the received encrypted character string in an otp storage space of the hardware. According to the hardware authorization method, the production date, the hardware serial number and the hardware model information of the hardware are adopted as machine codes, the machine codes can be linked with the production date and the production quantity, and the production quantity and the progress of the current date can be managed.

The invention relates to a defect detection device used for an air-blowing mini-type optical cable jacket surface. The defect detection device used for the air-blowing mini-type optical cable jacket surface comprises a detection seat sleeve. The detection seat sleeve is composed of an upper seat sleeve and a lower sleeve in a half-and-half opening and closing mode. Cable penetrating through holes are formed in the middles of the upper seat sleeve and the lower seat sleeve. One sides of the upper seat sleeve and the lower seat sleeve are connected in a hinged mode through a tensioner. Half-and-half sizing dies are arranged at the front ends and / or the rear ends of the upper seat sleeve and the lower seat sleeve in an installation mode. Contact alarm switches are arranged on the other sides of the upper seat sleeve and the lower seat sleeve in a configured mode. An aberration reflecting type detector is arranged in an installation mode on the circumference, along the cable penetrating through holes, of the detection seat sleeve. The defect detection device used for the air-blowing mini-type optical cable jacket surface has the advantages that due to the device is installed in a production process of a air-blowing mini-type optical cable, the whole segment length is detected, the defects of the optical cable surface can be effectively detected, detection precision is high, risks of air-blowing construction are avoided, due to adoption of the aberration reflecting type detector, the possible defects on each position of the air-blowing mini-type optical cable surface can be detected, precision is high, and therefore detection omission to small material dropping points of the mini-type optical cable is avoided. According to the detection seat sleeve, a structure of the upper seat sleeve and the lower seat sleeve in a half-and-half opening and closing mode is adopted, and use and installation are convenient to achieve.

The invention discloses a noise detection method and apparatus belonging to the technical field of terminals. The method comprises the following steps of: determining, for each frame signal of a plurality of frame signals included in a target audio signal, the number of subbands included in the frame signal and the frequency domain amplitude average value of each subband based on the cutoff frequency of the frame signal, wherein the target audio signal is an audio signal under noise detection; determining a noise detection coefficient of the target audio signal based on a frequency domain amplitude average value of each of the plurality of frame signals; and when the noise detection coefficient is set at a preset threshold interval, determining that the target audio signal is all noise. According to the method and apparatus of the invention, according to the noise detection coefficient determined by the frequency domain average value of the plurality of subbands of each of the plurality of frame signals, the target audio signal is determined to be all noise when the noise detection coefficient is within the preset coefficient threshold interval. The noise detection method and apparatus of the invention do not need operation intervention by working personnel so as to prevent occurrence of missed detection and accordingly raise the noise detection efficiency.

The invention discloses a detecting method of object position applied in circuit layout, which comprises capturing the coordinate range of the word objects of the circuit layout, then capturing the coordinate range of a to-be-compared object of the circuit layout, then detecting whether the coordinate range of the word objects superposes with the coordinate range of the to-be-compared object or the distance between both is smaller than the preset value. The method improves the human detecting mode adopted former, thereby shortening the detecting time of the layout, further increasing the quality of printed circuit boards.

The invention discloses a fatigue driving detection system based on vehicle networking. The fatigue driving detection system based on vehicle networking comprises a pressure sensor, a setting module,a mileage recording module, a lane deviation module, a central processing unit and a vehicle networking server, wherein the output end of the pressure sensor is connected with the input end of a forcesensor; the output end of the setting module is connected with the input end of a force module; and the output end of the force module is connected with the input end of a force comparison module. The invention relates to the technical field of fatigue driving. According to the fatigue driving detection system based on the vehicle networking, the pressure sensor can be used for sensing the grip force between the hands of a driver and a steering wheel, the grip force can be compared with the force set before in cooperation of the force comparison module, whether the grip force of a user is normal or not is detected, cooperation detection is carried out in cooperation with a brain wave sensor to give an alarm when in danger, fatigue driving is detected by the system from multiple angles atthe same time, and the problem of missing detection of the detection system is effectively avoided.

The invention belongs to the field of semiconductors, and especially relates to a method for screening out the abnormal crystal grains through the photoelectric property parameters of the crystal grains. The method achieves the screening of the abnormal crystal grains through the testing of the photoelectric properties of the crystal grains and a method of coordinate conversion of crystal grains, improves the work efficiency of a crystal grain mirror, reduces the detection leakage, reduces the complaint rate of customers, and improves the work efficiency.

The invention provides a chip detection and packaging production system, and relates to the technical field of chip manufacturing. The chip detection and packaging production system comprises a voltage withstand test device, an electrical test device, a chip pin detection device and an integrated device which are sequentially installed on a working platform, wherein the voltage withstand test device is used for carrying out voltage withstand detection on a chip; the electrical test device is used for carrying out electrical detection on the chip; the chip pin detection device is used for detecting the flatness, the intervals and the widths of chip pins; the integrated device is used for printing and packaging the chip; a first discharging conveyer belt of the voltage withstand test devicecommunicates with a feeding mechanism of a first feeding assembly of the electrical test device; a good product transferring mechanism of the electrical test device communicates with a second feedingassembly of the chip pin detection device; and a conveying mechanism of a second defective product collecting assembly of the chip pin detection device communicates with a conveying device of a printing assembly of the integrated device. According to the chip detection and packaging production system, the design is reasonable, the chip can be detected automatically and continuously, the labor investment is reduced, the detection omissions are avoided, and the chip quality is stable.

The invention discloses a target detection method for a three-coordinate active phased array radar. By respectively constructing a fast picture unit and a slow picture unit in different division manners for different distance ranges and respectively setting the corresponding threshold number of points for different picture units, when calculating the amplitude value threshold of a single picture unit in multi-frame accumulation, not only the amplitude value threshold of the single picture unit in the previous frame is considered, but also the number of points occupied by the echo data of the single picture unit within a certain number of frames is comprehensively considered to calculate and adjust the amplitude value threshold of the single picture unit. The method provided by the invention can effectively suppress the clutter and improve the detection capability of the target.

The invention discloses a jig for detecting a short-circuited or open-circuit component on a circuit board. The jig for detecting the short-circuited or open-circuit component on the circuit board comprises a pin plate, a pin protection plate, a roof plate, a press plate, PCB press bars, press bars, guide rods, a front plate and side plates, wherein the side plates which are perpendicularly vertical are arranged on the two sides of the lower plane of the pin plate, the front plate which is perpendicularly vertical is arranged on the front side of the lower plane of the pin plate, the two ends of the front plate are connected to the side plates, the pin protection plate is arranged on the pin plate, the multiple guide rods penetrating through the pin protection plate are arranged on the pin plate, the upper ends of the guide rods are connected with the press plate, the PCB press bars are arranged between the press plate and the pin protection plate, and the roof plate is installed on the upper plane of the press plate through the press bars. Due to the mode, detection efficiency of workers on an assembly line can be improved, detection omission is prevented, whether the short-circuited or open-circuit component is broken or not is completely judged through a detection program, and detection efficiency is greatly improved.

The invention discloses a laser correlation type computer mouse walking maze competition scoring system, and belongs to the field of communication. The system comprises a starting point signal detection device placed at the starting point of the maze, an end point signal detection device placed at the end point of the maze, a signal receiving device for monitoring the starting point signal and theend point signal, and upper computer scoring software for processing the starting point signal and the end point signal. The starting point signal detection device comprises a starting point laser emission module and a starting point illuminance detection module which are oppositely arranged; the end point signal detection device comprises an end point laser emission module and an end point illuminance detection module which are oppositely arranged; a starting point signal and an end point signal are detected by adopting a principle that a computer mouse shields laser beams emitted by a laseremission module to cause illuminance change of an illuminance detection module. Laser beams emitted by the laser emitting module are high in directivity, interference generated by light beam scattering is avoided, and therefore the accuracy of the scoring system is improved. The illuminance sensing module is high in detection speed, and the response speed of the scoring system is increased.

The invention discloses a detection system and a detection method for accurately positioning the leakage direction of a condenser titanium pipe. A detection belt fixing mechanism comprises a clampingplate, the inner side of the clamping plate is provided with a clamping groove for placing a detection belt, and the parts, located at the front side and the rear side of the clamping groove, of the clamping plate are rotatably connected with pressing wheels. The invention relates to the technical field of condensers. According to the detection system and the detection method for accurately positioning the leakage position of a condenser titanium pipe, the detection belt is rolled up to sleeve a pipeline and is fixed by the detection belt fixing mechanism, when the detection belt is pulled topass through the surface of the pipeline, the pipeline can be comprehensively detected, whether there is a leakage position or not is judged, if gas leaks, powder flies out of the surface of the detection belt, the leakage position can be visually seen, marking is convenient, positioning is accurate, the detection belt is fixed conveniently through the detection belt fixing mechanism, and the system can adapt to pipelines with different diameters has no limit on the variety of leaked gas, and is of high universality.

The invention discloses a Web vulnerability detection method and device, and the method comprises the steps: receiving a first instruction input by a user, running a preset script file based on the first instruction, and determining a legal IP in a preset IP file; based on a second instruction in the script file, controlling network connection end scanning software Nmap to run, and determining a uniform resource locator URL of each port which corresponds to the legal IP and provides the Web service; filtering the URL based on a preset URL white list, and generating a URL file based on the filtered URL; based on a third instruction preset in the script file, controlling network vulnerability scanning software AWVS to call the URL file, conducting vulnerability detection on each URL in the URL file, and generating a detection report is generated. The technical problem that in the prior art, the Web vulnerability detection efficiency is low is solved.

The invention discloses an elevator hand-pinch preventing method. The elevator hand-pinch preventing method comprises the following steps that S1, a camera obtains video images; S2, the video images are pretreated to obtain target area images of an elevator door; S3, the target area images are input into a hand-pinch preventing model, and through learning, a predicting result having the hand or having no hand is obtained; and S4, a corresponding action is conducted according to the predicting result. According to the elevator hand-pinch preventing method, by shooting the video images, having the hand or having no hand is predicted on the images on the edge of the elevator door, thus existence of small objects at the edge of the elevator door is judged, and the probability of accidents is lowered.

The invention discloses a vehicle detection and tracking method and device, an apparatus and a medium. The method comprises the following steps: extracting a feature descriptor of an image of a vehicle to be tracked; detecting A target vehicle tracking object in an image of a region to be detected by using a trained vehicle detection model including features of a vehicle normally visible region component and feature of a vehicle easily occluded region component; Extracting feature descriptors of target vehicle tracking object images; performing Local sensitive hash matching and purification of feature descriptors; Determining the number of feature descriptors after purification and judging whether the number is larger than a preset threshold value; If so, the target vehicle tracking object is tracked when the area hue value of the target vehicle tracking object is within a preset hue value range. That is, the present invention utilizes a model including a common visible area and a easily occluded area to perform image detection, detects a vehicle to be tracked through a local sensitive hash matching, realizes vehicle tracking, effectively avoids the problem of vehicle leakage detection due to occlusion, and improves tracking accuracy.

The embodiment of the invention discloses a memorizer as well as a detecting method thereof and a chip. The memorizer comprises a substrate, a laser detection circuit located on the substrate as well as a data circuit area and a data storage area located on the substrate; the laser detection circuit at least comprises a plurality of monitoring diodes, at least one recharge transistor and at least one amplifier unit, the input end of each amplifier unit is electrically connected with the negative electrode of at least one monitoring diode and the output end of each recharge transistor respectively, the positive electrodes of the monitoring diodes are grounded, and the input end of each recharge transistor is electrically connected with a power supply; the data circuit area is provided with a plurality of sensitive circuits, the data storage area is provided with a storage array, each sensitive circuit is arranged adjacent to at least one monitoring diode, and / or the storage array is arranged adjacent to at least one monitoring diode. The memorizer provided by the embodiment of the invention is low in production cost, is capable of effectively improving the accuracy of laser detection and avoiding miss detection.

The invention discloses a product inspection machine and a control method thereof, and relates to the technical field of detection devices. The product inspection machine comprises a detection platform, a waste discharging platform, a discharging platform, a main control system and an air suction conveying belt; the detection platform comprises a feeding mechanism, a first sensing device and a visual scanning device which are arranged on the detection platform in sequence, and the visual scanning device is used for shooting and detecting; and the waste discharging platform comprises a blowingdevice and a second sensing device which are sequentially arranged on the waste discharging platform, the discharging platform comprises a discharging mechanism, a gap is formed between the detectionplatform and the waste discharge platform, the air blowing port of the air blowing device corresponds to a notch, a waste collecting box is arranged at the bottom of the waste discharging platform, and the waste collecting box corresponds to the notch. Materials are conveyed through the air suction conveying belt, the detection platform and the main control system are used for detecting the materials, waste materials are discharged by the waste discharging platform, and the discharging platform is used for discharging, so that the inspection machine does not need to be manually detected, and the detection efficiency and the detection precision are improved.

The invention discloses an omni-directional product appearance detection device. The omni-directional product appearance detection device comprises a conveying belt, a rotating motor, a fixed ring andelectric push rods, wherein a storage tray is fixedly arranged on the upper end surface of the conveying belt; a detection camera and an indicating lamp are arranged on the inner sides of side platesrespectively; the rotating motor is arranged on a support, and the transmission end of the rotating motor is vertically connected with a rotating disc; a ball is arranged between the storage tray andthe conveying belt; the edge of the upper end surface of the fixed ring is vertically connected with one end of a positioning rod; the electric push rods are fixedly arranged on the outer sides of the side plates; printing cotton is embedded in a connecting piece; liquid storage boxes are fixed on the top of the outer sides of the side plates; an iron core is fixedly arranged in an isolation shell, the outer side of the iron core is wrapped with a wire; and a control center is mounted on the outer side of the conveying belt. By virtue of the omni-directional product appearance detection device, the detection omission problem which possibly occurs in manual detection is effectively avoided, and the unqualified products can be classified.